diff --git "a/results.json" "b/results.json"
--- "a/results.json"
+++ "b/results.json"
@@ -8,7 +8,7 @@
       "family": "Indo-European",
       "flores_path": "eng_Latn",
       "fleurs_tag": "en_us",
-      "commonvoice_hours": 2659.0,
+      "commonvoice_hours": 2660.0,
       "commonvoice_locale": "en",
       "in_benchmark": true
     },
@@ -80,7 +80,7 @@
       "family": "Indo-European",
       "flores_path": "fra_Latn",
       "fleurs_tag": "fr_fr",
-      "commonvoice_hours": 1055.0,
+      "commonvoice_hours": 1056.0,
       "commonvoice_locale": "fr",
       "in_benchmark": true
     },
@@ -164,7 +164,7 @@
       "family": "Indo-European",
       "flores_path": "deu_Latn",
       "fleurs_tag": "de_de",
-      "commonvoice_hours": 1361.0,
+      "commonvoice_hours": 1362.0,
       "commonvoice_locale": "de",
       "in_benchmark": true
     },
@@ -248,7 +248,7 @@
       "family": "Dravidian",
       "flores_path": "tam_Taml",
       "fleurs_tag": "ta_in",
-      "commonvoice_hours": 234.0,
+      "commonvoice_hours": 235.0,
       "commonvoice_locale": "ta",
       "in_benchmark": true
     },
@@ -1292,7 +1292,7 @@
       "family": "Indo-European",
       "flores_path": "cat_Latn",
       "fleurs_tag": "ca_es",
-      "commonvoice_hours": 2851.0,
+      "commonvoice_hours": 2854.0,
       "commonvoice_locale": "ca",
       "in_benchmark": true
     },
@@ -1304,7 +1304,7 @@
       "family": "Afro-Asiatic",
       "flores_path": "heb_Hebr",
       "fleurs_tag": "he_il",
-      "commonvoice_hours": 1.1,
+      "commonvoice_hours": 1.2,
       "commonvoice_locale": "he",
       "in_benchmark": true
     },
@@ -1376,7 +1376,7 @@
       "family": "Turkic",
       "flores_path": "uig_Arab",
       "fleurs_tag": null,
-      "commonvoice_hours": 367.0,
+      "commonvoice_hours": 368.0,
       "commonvoice_locale": "ug",
       "in_benchmark": true
     },
@@ -1748,7 +1748,7 @@
       "family": "Indo-European",
       "flores_path": "nob_Latn",
       "fleurs_tag": "nb_no",
-      "commonvoice_hours": 0.1,
+      "commonvoice_hours": 0.3,
       "commonvoice_locale": "nb-NO",
       "in_benchmark": true
     },
@@ -1832,7 +1832,7 @@
       "family": "Tai-Kadai",
       "flores_path": "lao_Laoo",
       "fleurs_tag": "lo_la",
-      "commonvoice_hours": 0.2,
+      "commonvoice_hours": 0.3,
       "commonvoice_locale": "lo",
       "in_benchmark": true
     },
@@ -2156,7 +2156,7 @@
       "family": "Kartvelian",
       "flores_path": "kat_Geor",
       "fleurs_tag": "ka_ge",
-      "commonvoice_hours": 163.0,
+      "commonvoice_hours": 164.0,
       "commonvoice_locale": "ka",
       "in_benchmark": true
     },
@@ -2168,7 +2168,7 @@
       "family": "Indo-European",
       "flores_path": "glg_Latn",
       "fleurs_tag": "gl_es",
-      "commonvoice_hours": 114.0,
+      "commonvoice_hours": 115.0,
       "commonvoice_locale": "gl",
       "in_benchmark": true
     },
@@ -2336,7 +2336,7 @@
       "family": "Atlantic-Congo",
       "flores_path": null,
       "fleurs_tag": null,
-      "commonvoice_hours": 2.0,
+      "commonvoice_hours": 2.1,
       "commonvoice_locale": "bci",
       "in_benchmark": false
     },
@@ -2432,8 +2432,8 @@
       "family": "Indo-European",
       "flores_path": null,
       "fleurs_tag": null,
-      "commonvoice_hours": null,
-      "commonvoice_locale": null,
+      "commonvoice_hours": 0.0,
+      "commonvoice_locale": "jam",
       "in_benchmark": false
     },
     {
@@ -3536,7 +3536,7 @@
       "family": null,
       "flores_path": "eus_Latn",
       "fleurs_tag": null,
-      "commonvoice_hours": 336.0,
+      "commonvoice_hours": 337.0,
       "commonvoice_locale": "eu",
       "in_benchmark": true
     },
@@ -3560,7 +3560,7 @@
       "family": "Abkhaz-Adyge",
       "flores_path": null,
       "fleurs_tag": null,
-      "commonvoice_hours": 38.0,
+      "commonvoice_hours": 44.0,
       "commonvoice_locale": "kbd",
       "in_benchmark": false
     },
@@ -4352,7 +4352,7 @@
       "family": "Indo-European",
       "flores_path": null,
       "fleurs_tag": null,
-      "commonvoice_hours": 28.0,
+      "commonvoice_hours": 29.0,
       "commonvoice_locale": "br",
       "in_benchmark": false
     },
@@ -4976,7 +4976,7 @@
       "family": "Indo-European",
       "flores_path": null,
       "fleurs_tag": null,
-      "commonvoice_hours": 0.6,
+      "commonvoice_hours": 4.8,
       "commonvoice_locale": "kvx",
       "in_benchmark": false
     },
@@ -5420,7 +5420,7 @@
       "family": "Indo-European",
       "flores_path": null,
       "fleurs_tag": null,
-      "commonvoice_hours": 0.2,
+      "commonvoice_hours": 2.7,
       "commonvoice_locale": "kxp",
       "in_benchmark": false
     },
@@ -5432,7 +5432,7 @@
       "family": "Indo-European",
       "flores_path": null,
       "fleurs_tag": null,
-      "commonvoice_hours": 1.1,
+      "commonvoice_hours": 11.0,
       "commonvoice_locale": "gjk",
       "in_benchmark": false
     },
@@ -8142,7 +8142,7 @@
       "id": "meta-llama/llama-4-maverick",
       "name": "Llama 4 Maverick",
       "provider_name": "Meta",
-      "cost": 0.85,
+      "cost": 0.6,
       "hf_id": "meta-llama/Llama-4-Maverick-17B-128E-Instruct",
       "size": 401583781376.0,
       "type": "Open",
@@ -8153,7 +8153,7 @@
       "id": "meta-llama/llama-3.3-70b-instruct",
       "name": "Llama 3.3 70B Instruct",
       "provider_name": "Meta",
-      "cost": 0.3,
+      "cost": 0.25,
       "hf_id": "meta-llama/Llama-3.3-70B-Instruct",
       "size": 70553706496.0,
       "type": "Open",
@@ -8164,7 +8164,7 @@
       "id": "meta-llama/llama-3.1-70b-instruct",
       "name": "Llama 3.1 70B Instruct",
       "provider_name": "Meta",
-      "cost": 0.3,
+      "cost": 0.28,
       "hf_id": "meta-llama/Llama-3.1-70B-Instruct",
       "size": 70553706496.0,
       "type": "Open",
@@ -8182,6 +8182,17 @@
       "license": "Llama3",
       "creation_date": "2024-04-17"
     },
+    {
+      "id": "openai/gpt-4.1-nano",
+      "name": "GPT-4.1 Nano",
+      "provider_name": "OpenAI",
+      "cost": 0.4,
+      "hf_id": null,
+      "size": null,
+      "type": "Commercial",
+      "license": null,
+      "creation_date": "2025-04-14"
+    },
     {
       "id": "openai/gpt-4o-mini",
       "name": "GPT-4o-mini",
@@ -8204,6 +8215,17 @@
       "license": "Apache 2.0",
       "creation_date": "2025-03-11"
     },
+    {
+      "id": "google/gemini-2.5-flash-preview",
+      "name": "Gemini 2.5 Flash Preview",
+      "provider_name": "Google",
+      "cost": 0.6,
+      "hf_id": null,
+      "size": null,
+      "type": "Commercial",
+      "license": null,
+      "creation_date": "2025-04-17"
+    },
     {
       "id": "google/gemma-3-27b-it",
       "name": "Gemma 3 27B",
@@ -8325,6 +8347,22 @@
       "score": 0.5835344719191324,
       "sentence_nr": 0
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "en",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.5068360865084963,
+      "sentence_nr": 0
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "en",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.6984477545886919,
+      "sentence_nr": 0
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "en",
@@ -8357,6 +8395,22 @@
       "score": 0.5880210095195896,
       "sentence_nr": 0
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "en",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.5767974034153417,
+      "sentence_nr": 0
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "en",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.7416604155998838,
+      "sentence_nr": 0
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "en",
@@ -8501,6 +8555,22 @@
       "score": 0.5246291817407542,
       "sentence_nr": 0
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "zh",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.19685996214444484,
+      "sentence_nr": 0
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "zh",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.45295467854325566,
+      "sentence_nr": 0
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "zh",
@@ -8533,6 +8603,22 @@
       "score": 0.5164808837319497,
       "sentence_nr": 0
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "zh",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.39544143038459856,
+      "sentence_nr": 0
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "zh",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.5959322291414695,
+      "sentence_nr": 0
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "zh",
@@ -8677,6 +8763,22 @@
       "score": 0.9463396364218181,
       "sentence_nr": 0
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "hi",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.8347630061919914,
+      "sentence_nr": 0
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "hi",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.90120785919445,
+      "sentence_nr": 0
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "hi",
@@ -8709,6 +8811,22 @@
       "score": 0.9958930217841712,
       "sentence_nr": 0
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "hi",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.9878765474230741,
+      "sentence_nr": 0
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "hi",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.9958930217841712,
+      "sentence_nr": 0
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "hi",
@@ -8853,6 +8971,22 @@
       "score": 0.4820043660869366,
       "sentence_nr": 0
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "es",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.2755180390863056,
+      "sentence_nr": 0
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "es",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.45513546372451297,
+      "sentence_nr": 0
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "es",
@@ -8885,6 +9019,22 @@
       "score": 0.49819657249183386,
       "sentence_nr": 0
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "es",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.498027570572246,
+      "sentence_nr": 0
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "es",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.6323394518481261,
+      "sentence_nr": 0
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "es",
@@ -9029,6 +9179,22 @@
       "score": 0.5861956606716949,
       "sentence_nr": 0
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ar",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.3688167255318499,
+      "sentence_nr": 0
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ar",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.5626506837530694,
+      "sentence_nr": 0
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ar",
@@ -9061,6 +9227,22 @@
       "score": 0.5946895227088745,
       "sentence_nr": 0
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ar",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.5946872253540679,
+      "sentence_nr": 0
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ar",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.7456034598536061,
+      "sentence_nr": 0
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ar",
@@ -9205,6 +9387,22 @@
       "score": 0.5666753970394321,
       "sentence_nr": 0
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ur",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.30290762996936516,
+      "sentence_nr": 0
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ur",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.5118703718099534,
+      "sentence_nr": 0
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ur",
@@ -9237,6 +9435,22 @@
       "score": 0.4681164293806726,
       "sentence_nr": 0
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ur",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.46238620976190176,
+      "sentence_nr": 0
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ur",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.6294498584668344,
+      "sentence_nr": 0
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ur",
@@ -9381,6 +9595,22 @@
       "score": 0.4540589962283635,
       "sentence_nr": 0
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "fr",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.3146876332155584,
+      "sentence_nr": 0
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "fr",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.5132307789796908,
+      "sentence_nr": 0
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "fr",
@@ -9413,6 +9643,22 @@
       "score": 0.4892722276483434,
       "sentence_nr": 0
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "fr",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.3968190443102714,
+      "sentence_nr": 0
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "fr",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.5978315904237319,
+      "sentence_nr": 0
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "fr",
@@ -9557,6 +9803,22 @@
       "score": 0.5834523243646894,
       "sentence_nr": 0
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "bn",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.4116786321664925,
+      "sentence_nr": 0
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "bn",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.5837404006033683,
+      "sentence_nr": 0
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "bn",
@@ -9589,6 +9851,22 @@
       "score": 0.5861999156017297,
       "sentence_nr": 0
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "bn",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.4692988665122676,
+      "sentence_nr": 0
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "bn",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.6232055062803948,
+      "sentence_nr": 0
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "bn",
@@ -9733,6 +10011,22 @@
       "score": 0.5344680037267059,
       "sentence_nr": 0
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pt",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.38857454128250113,
+      "sentence_nr": 0
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pt",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.568493737147045,
+      "sentence_nr": 0
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "pt",
@@ -9765,6 +10059,22 @@
       "score": 0.5432217848942439,
       "sentence_nr": 0
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pt",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.5815914780944419,
+      "sentence_nr": 0
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pt",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.7321831383933567,
+      "sentence_nr": 0
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "pt",
@@ -9909,6 +10219,22 @@
       "score": 0.6547171931962555,
       "sentence_nr": 0
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pa",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.39116961050753873,
+      "sentence_nr": 0
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pa",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.6565815107523159,
+      "sentence_nr": 0
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "pa",
@@ -9941,6 +10267,22 @@
       "score": 0.6533822343227146,
       "sentence_nr": 0
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pa",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.45761264167159527,
+      "sentence_nr": 0
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pa",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.6880668163480468,
+      "sentence_nr": 0
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "pa",
@@ -10085,6 +10427,22 @@
       "score": 0.5465517653500693,
       "sentence_nr": 0
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ru",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.33599052457705525,
+      "sentence_nr": 0
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ru",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.5506683357790431,
+      "sentence_nr": 0
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ru",
@@ -10117,6 +10475,22 @@
       "score": 0.5410881356746259,
       "sentence_nr": 0
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ru",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.5444545010896156,
+      "sentence_nr": 0
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ru",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.6995635476507981,
+      "sentence_nr": 0
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ru",
@@ -10261,6 +10635,22 @@
       "score": 0.4110516731487298,
       "sentence_nr": 0
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "sw",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.3497458686123802,
+      "sentence_nr": 0
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "sw",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.5486997385559668,
+      "sentence_nr": 0
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "sw",
@@ -10293,6 +10683,22 @@
       "score": 0.5497542561082874,
       "sentence_nr": 0
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "sw",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.5116576338831017,
+      "sentence_nr": 0
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "sw",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.6761326472499163,
+      "sentence_nr": 0
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "sw",
@@ -10437,6 +10843,22 @@
       "score": 0.6608358312257032,
       "sentence_nr": 0
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "id",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.41109781447793037,
+      "sentence_nr": 0
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "id",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.5909828034438983,
+      "sentence_nr": 0
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "id",
@@ -10469,6 +10891,22 @@
       "score": 0.6283965584123504,
       "sentence_nr": 0
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "id",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.4805240060295435,
+      "sentence_nr": 0
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "id",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.6678968008486404,
+      "sentence_nr": 0
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "id",
@@ -10613,6 +11051,22 @@
       "score": 0.6326418045965277,
       "sentence_nr": 0
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "de",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.464941535335407,
+      "sentence_nr": 0
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "de",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.6614464392132021,
+      "sentence_nr": 0
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "de",
@@ -10645,6 +11099,22 @@
       "score": 0.681786235656136,
       "sentence_nr": 0
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "de",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.5963279752582172,
+      "sentence_nr": 0
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "de",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.7384834127486712,
+      "sentence_nr": 0
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "de",
@@ -10789,6 +11259,22 @@
       "score": 0.42969616197156246,
       "sentence_nr": 0
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ja",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.19979408064496967,
+      "sentence_nr": 0
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ja",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.44501420508709394,
+      "sentence_nr": 0
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ja",
@@ -10821,6 +11307,22 @@
       "score": 0.4522509933949415,
       "sentence_nr": 0
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ja",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.4278755948981275,
+      "sentence_nr": 0
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ja",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.6049024929611014,
+      "sentence_nr": 0
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ja",
@@ -10965,6 +11467,22 @@
       "score": 0.46822754470803873,
       "sentence_nr": 1
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "en",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.0,
+      "sentence_nr": 1
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "en",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.4463892736716987,
+      "sentence_nr": 1
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "en",
@@ -10997,6 +11515,22 @@
       "score": 0.4031456247133876,
       "sentence_nr": 1
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "en",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.1583163321869008,
+      "sentence_nr": 1
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "en",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.5032511267062394,
+      "sentence_nr": 1
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "en",
@@ -11141,6 +11675,22 @@
       "score": 0.4427230465401631,
       "sentence_nr": 1
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "zh",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.05534265631745826,
+      "sentence_nr": 1
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "zh",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.39688946206212833,
+      "sentence_nr": 1
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "zh",
@@ -11173,6 +11723,22 @@
       "score": 0.38435741328258305,
       "sentence_nr": 1
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "zh",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.0,
+      "sentence_nr": 1
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "zh",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.42954139521687473,
+      "sentence_nr": 1
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "zh",
@@ -11317,6 +11883,22 @@
       "score": 0.3356633416447032,
       "sentence_nr": 1
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "hi",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.09553543457519309,
+      "sentence_nr": 1
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "hi",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.3531525294256142,
+      "sentence_nr": 1
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "hi",
@@ -11349,6 +11931,22 @@
       "score": 0.28789057461471257,
       "sentence_nr": 1
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "hi",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.1568418931847707,
+      "sentence_nr": 1
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "hi",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.41485110412488607,
+      "sentence_nr": 1
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "hi",
@@ -11493,6 +12091,22 @@
       "score": 0.47645148444499064,
       "sentence_nr": 1
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "es",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.0,
+      "sentence_nr": 1
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "es",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.4042977714165801,
+      "sentence_nr": 1
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "es",
@@ -11525,6 +12139,22 @@
       "score": 0.38785209659947417,
       "sentence_nr": 1
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "es",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.0,
+      "sentence_nr": 1
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "es",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.43122763125947844,
+      "sentence_nr": 1
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "es",
@@ -11669,6 +12299,22 @@
       "score": 0.5953162569846108,
       "sentence_nr": 1
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ar",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.22238763863547817,
+      "sentence_nr": 1
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ar",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.5765887803460186,
+      "sentence_nr": 1
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ar",
@@ -11701,6 +12347,22 @@
       "score": 0.5179253053631742,
       "sentence_nr": 1
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ar",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.23522101642407195,
+      "sentence_nr": 1
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ar",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.5302138314227511,
+      "sentence_nr": 1
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ar",
@@ -11845,6 +12507,22 @@
       "score": 0.3833939462124923,
       "sentence_nr": 1
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ur",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.06036796843527163,
+      "sentence_nr": 1
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ur",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.37581426760977427,
+      "sentence_nr": 1
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ur",
@@ -11877,6 +12555,22 @@
       "score": 0.2718653389257641,
       "sentence_nr": 1
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ur",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.0,
+      "sentence_nr": 1
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ur",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.4551974335055929,
+      "sentence_nr": 1
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ur",
@@ -12021,6 +12715,22 @@
       "score": 0.41747276065817185,
       "sentence_nr": 1
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "fr",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.0,
+      "sentence_nr": 1
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "fr",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.401865675252717,
+      "sentence_nr": 1
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "fr",
@@ -12053,6 +12763,22 @@
       "score": 0.3853293582383978,
       "sentence_nr": 1
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "fr",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.08025555322499443,
+      "sentence_nr": 1
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "fr",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.40116594181297777,
+      "sentence_nr": 1
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "fr",
@@ -12197,6 +12923,22 @@
       "score": 0.3829169125379508,
       "sentence_nr": 1
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "bn",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.0,
+      "sentence_nr": 1
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "bn",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.3404791678264965,
+      "sentence_nr": 1
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "bn",
@@ -12229,6 +12971,22 @@
       "score": 0.335890201952113,
       "sentence_nr": 1
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "bn",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.08465714266003518,
+      "sentence_nr": 1
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "bn",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.41240280272579144,
+      "sentence_nr": 1
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "bn",
@@ -12373,6 +13131,22 @@
       "score": 0.4075406301092705,
       "sentence_nr": 1
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pt",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.09107675218561961,
+      "sentence_nr": 1
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pt",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.4297036775694859,
+      "sentence_nr": 1
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "pt",
@@ -12405,6 +13179,22 @@
       "score": 0.3960585990192623,
       "sentence_nr": 1
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pt",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.1262345212021199,
+      "sentence_nr": 1
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pt",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.4305259421555756,
+      "sentence_nr": 1
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "pt",
@@ -12549,6 +13339,22 @@
       "score": 0.3714280466838255,
       "sentence_nr": 1
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pa",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.10639077953820089,
+      "sentence_nr": 1
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pa",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.44452351549730684,
+      "sentence_nr": 1
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "pa",
@@ -12581,6 +13387,22 @@
       "score": 0.3538602132402044,
       "sentence_nr": 1
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pa",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.09058017202418268,
+      "sentence_nr": 1
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pa",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.45761691752578293,
+      "sentence_nr": 1
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "pa",
@@ -12725,6 +13547,22 @@
       "score": 0.45371814600333005,
       "sentence_nr": 1
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ru",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.0,
+      "sentence_nr": 1
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ru",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.4351068027997583,
+      "sentence_nr": 1
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ru",
@@ -12757,6 +13595,22 @@
       "score": 0.3980589439671235,
       "sentence_nr": 1
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ru",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.0,
+      "sentence_nr": 1
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ru",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.4357976697626214,
+      "sentence_nr": 1
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ru",
@@ -12901,6 +13755,22 @@
       "score": 0.34155562837143877,
       "sentence_nr": 1
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "sw",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.07977475994840084,
+      "sentence_nr": 1
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "sw",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.41362647590171586,
+      "sentence_nr": 1
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "sw",
@@ -12933,6 +13803,22 @@
       "score": 0.1139393935967296,
       "sentence_nr": 1
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "sw",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.12405768590988119,
+      "sentence_nr": 1
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "sw",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.40898021337964235,
+      "sentence_nr": 1
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "sw",
@@ -13077,6 +13963,22 @@
       "score": 0.41031664319131844,
       "sentence_nr": 1
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "id",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.0,
+      "sentence_nr": 1
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "id",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.3867446668969619,
+      "sentence_nr": 1
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "id",
@@ -13109,6 +14011,22 @@
       "score": 0.2626949949898101,
       "sentence_nr": 1
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "id",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.0,
+      "sentence_nr": 1
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "id",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.4345822107320723,
+      "sentence_nr": 1
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "id",
@@ -13253,6 +14171,22 @@
       "score": 0.3938156291645021,
       "sentence_nr": 1
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "de",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.0,
+      "sentence_nr": 1
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "de",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.4423346652606821,
+      "sentence_nr": 1
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "de",
@@ -13285,6 +14219,22 @@
       "score": 0.4098374118843212,
       "sentence_nr": 1
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "de",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.0,
+      "sentence_nr": 1
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "de",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.4537465621062763,
+      "sentence_nr": 1
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "de",
@@ -13429,6 +14379,22 @@
       "score": 0.3889987132692464,
       "sentence_nr": 1
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ja",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.06058139791899572,
+      "sentence_nr": 1
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ja",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.4408463415099943,
+      "sentence_nr": 1
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ja",
@@ -13461,6 +14427,22 @@
       "score": 0.3916082207331212,
       "sentence_nr": 1
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ja",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.08841818041776058,
+      "sentence_nr": 1
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ja",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.47772386173878106,
+      "sentence_nr": 1
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ja",
@@ -13605,6 +14587,22 @@
       "score": 0.7037574715738644,
       "sentence_nr": 2
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "en",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.44323526910431466,
+      "sentence_nr": 2
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "en",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.7203175886481126,
+      "sentence_nr": 2
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "en",
@@ -13637,6 +14635,22 @@
       "score": 0.49051792813181655,
       "sentence_nr": 2
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "en",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.005449161724399305,
+      "sentence_nr": 2
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "en",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.026158029267484995,
+      "sentence_nr": 2
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "en",
@@ -13781,6 +14795,22 @@
       "score": 0.31139762378406344,
       "sentence_nr": 2
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "zh",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.06724888422961112,
+      "sentence_nr": 2
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "zh",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.39667480605700844,
+      "sentence_nr": 2
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "zh",
@@ -13813,6 +14843,22 @@
       "score": 0.3675317022605926,
       "sentence_nr": 2
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "zh",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.2377604053257556,
+      "sentence_nr": 2
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "zh",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.5690390533910819,
+      "sentence_nr": 2
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "zh",
@@ -13957,6 +15003,22 @@
       "score": 0.5084057058209687,
       "sentence_nr": 2
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "hi",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.2868985878682555,
+      "sentence_nr": 2
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "hi",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.591501744009396,
+      "sentence_nr": 2
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "hi",
@@ -13989,6 +15051,22 @@
       "score": 0.063050817196087,
       "sentence_nr": 2
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "hi",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.24342570806900707,
+      "sentence_nr": 2
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "hi",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.5341782261409304,
+      "sentence_nr": 2
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "hi",
@@ -14133,6 +15211,22 @@
       "score": 0.6344846206551544,
       "sentence_nr": 2
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "es",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.3428955163829333,
+      "sentence_nr": 2
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "es",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.6023036718160529,
+      "sentence_nr": 2
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "es",
@@ -14165,6 +15259,22 @@
       "score": 0.501302719796297,
       "sentence_nr": 2
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "es",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.4538010737057216,
+      "sentence_nr": 2
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "es",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.6446834621229663,
+      "sentence_nr": 2
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "es",
@@ -14309,6 +15419,22 @@
       "score": 0.55968513851572,
       "sentence_nr": 2
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ar",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.29037747307996287,
+      "sentence_nr": 2
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ar",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.5917006930610393,
+      "sentence_nr": 2
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ar",
@@ -14341,6 +15467,22 @@
       "score": 0.39152357647177133,
       "sentence_nr": 2
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ar",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.4054234087021839,
+      "sentence_nr": 2
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ar",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.667304751638097,
+      "sentence_nr": 2
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ar",
@@ -14485,6 +15627,22 @@
       "score": 0.4646043403137081,
       "sentence_nr": 2
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ur",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.12049505059461789,
+      "sentence_nr": 2
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ur",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.49305039430054654,
+      "sentence_nr": 2
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ur",
@@ -14517,6 +15675,22 @@
       "score": 0.13083094614009624,
       "sentence_nr": 2
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ur",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.30865045220428267,
+      "sentence_nr": 2
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ur",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.61517480898171,
+      "sentence_nr": 2
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ur",
@@ -14661,6 +15835,22 @@
       "score": 0.6496192656497308,
       "sentence_nr": 2
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "fr",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.36781689904382464,
+      "sentence_nr": 2
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "fr",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.6454684777803729,
+      "sentence_nr": 2
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "fr",
@@ -14693,6 +15883,22 @@
       "score": 0.4799723286048352,
       "sentence_nr": 2
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "fr",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.005449161724399305,
+      "sentence_nr": 2
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "fr",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.026158029267484995,
+      "sentence_nr": 2
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "fr",
@@ -14837,6 +16043,22 @@
       "score": 0.5067677916637257,
       "sentence_nr": 2
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "bn",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.0,
+      "sentence_nr": 2
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "bn",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.49804532928450235,
+      "sentence_nr": 2
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "bn",
@@ -14869,6 +16091,22 @@
       "score": 0.4228574070038002,
       "sentence_nr": 2
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "bn",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.39464291294535175,
+      "sentence_nr": 2
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "bn",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.6481030286156323,
+      "sentence_nr": 2
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "bn",
@@ -15013,6 +16251,22 @@
       "score": 0.6252821653079126,
       "sentence_nr": 2
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pt",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.30783677787322206,
+      "sentence_nr": 2
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pt",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.6417108947268295,
+      "sentence_nr": 2
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "pt",
@@ -15045,6 +16299,22 @@
       "score": 0.49793621556542356,
       "sentence_nr": 2
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pt",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.44464935391849836,
+      "sentence_nr": 2
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pt",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.696223947326307,
+      "sentence_nr": 2
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "pt",
@@ -15189,6 +16459,22 @@
       "score": 0.43592329727028295,
       "sentence_nr": 2
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pa",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.2162882016096312,
+      "sentence_nr": 2
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pa",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.5347607537943324,
+      "sentence_nr": 2
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "pa",
@@ -15221,6 +16507,22 @@
       "score": 0.3347576434758551,
       "sentence_nr": 2
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pa",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.37081839104772296,
+      "sentence_nr": 2
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pa",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.6563960892873841,
+      "sentence_nr": 2
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "pa",
@@ -15365,6 +16667,22 @@
       "score": 0.5098952451698188,
       "sentence_nr": 2
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ru",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.2537710754125116,
+      "sentence_nr": 2
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ru",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.5898314098579354,
+      "sentence_nr": 2
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ru",
@@ -15397,6 +16715,22 @@
       "score": 0.4275543759804943,
       "sentence_nr": 2
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ru",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.3319944964021059,
+      "sentence_nr": 2
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ru",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.6676204564640195,
+      "sentence_nr": 2
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ru",
@@ -15541,6 +16875,22 @@
       "score": 0.5368112087257564,
       "sentence_nr": 2
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "sw",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.31664648301486426,
+      "sentence_nr": 2
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "sw",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.5144731747994161,
+      "sentence_nr": 2
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "sw",
@@ -15573,6 +16923,22 @@
       "score": 0.37413906702142435,
       "sentence_nr": 2
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "sw",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.005449161724399305,
+      "sentence_nr": 2
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "sw",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.026158029267484995,
+      "sentence_nr": 2
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "sw",
@@ -15717,6 +17083,22 @@
       "score": 0.6338401824373191,
       "sentence_nr": 2
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "id",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.3842999367495742,
+      "sentence_nr": 2
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "id",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.5816504483384909,
+      "sentence_nr": 2
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "id",
@@ -15749,6 +17131,22 @@
       "score": 0.4938296655037709,
       "sentence_nr": 2
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "id",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.546590882357469,
+      "sentence_nr": 2
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "id",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.7387035187469114,
+      "sentence_nr": 2
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "id",
@@ -15893,6 +17291,22 @@
       "score": 0.5106509239874657,
       "sentence_nr": 2
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "de",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.2258043389079604,
+      "sentence_nr": 2
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "de",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.528195718512124,
+      "sentence_nr": 2
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "de",
@@ -15925,6 +17339,22 @@
       "score": 0.4522763055702811,
       "sentence_nr": 2
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "de",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.005449161724399305,
+      "sentence_nr": 2
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "de",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.026158029267484995,
+      "sentence_nr": 2
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "de",
@@ -16069,6 +17499,22 @@
       "score": 0.404377371664668,
       "sentence_nr": 2
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ja",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.2573392925035755,
+      "sentence_nr": 2
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ja",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.5328565784271402,
+      "sentence_nr": 2
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ja",
@@ -16101,6 +17547,22 @@
       "score": 0.44607340294350173,
       "sentence_nr": 2
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ja",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.005449161724399305,
+      "sentence_nr": 2
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ja",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.026158029267484995,
+      "sentence_nr": 2
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ja",
@@ -16245,6 +17707,22 @@
       "score": 0.5299556742893647,
       "sentence_nr": 3
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "en",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.369345079296433,
+      "sentence_nr": 3
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "en",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.4718547623527638,
+      "sentence_nr": 3
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "en",
@@ -16277,6 +17755,22 @@
       "score": 0.5103516764863386,
       "sentence_nr": 3
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "en",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.4646232199104102,
+      "sentence_nr": 3
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "en",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.5678926447384061,
+      "sentence_nr": 3
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "en",
@@ -16421,6 +17915,22 @@
       "score": 0.280413108453108,
       "sentence_nr": 3
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "zh",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.13857814312261593,
+      "sentence_nr": 3
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "zh",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.26136391784313634,
+      "sentence_nr": 3
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "zh",
@@ -16453,6 +17963,22 @@
       "score": 0.2510112235832054,
       "sentence_nr": 3
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "zh",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.1767874865365185,
+      "sentence_nr": 3
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "zh",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.31807700660641347,
+      "sentence_nr": 3
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "zh",
@@ -16597,6 +18123,22 @@
       "score": 0.5296624608564717,
       "sentence_nr": 3
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "hi",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.4272870063962341,
+      "sentence_nr": 3
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "hi",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.5425603129070803,
+      "sentence_nr": 3
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "hi",
@@ -16629,6 +18171,22 @@
       "score": 0.5402588602256685,
       "sentence_nr": 3
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "hi",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.6242817472465665,
+      "sentence_nr": 3
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "hi",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.6834390596430621,
+      "sentence_nr": 3
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "hi",
@@ -16773,6 +18331,22 @@
       "score": 0.44716007458096513,
       "sentence_nr": 3
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "es",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.27046570133003095,
+      "sentence_nr": 3
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "es",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.3736509638751225,
+      "sentence_nr": 3
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "es",
@@ -16805,6 +18379,22 @@
       "score": 0.4224404198283467,
       "sentence_nr": 3
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "es",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.30147856626075187,
+      "sentence_nr": 3
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "es",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.4883780556286986,
+      "sentence_nr": 3
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "es",
@@ -16949,6 +18539,22 @@
       "score": 0.30391153783979835,
       "sentence_nr": 3
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ar",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.12887696534828325,
+      "sentence_nr": 3
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ar",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.30114368429557287,
+      "sentence_nr": 3
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ar",
@@ -16981,6 +18587,22 @@
       "score": 0.3343587266874694,
       "sentence_nr": 3
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ar",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.43310177167002284,
+      "sentence_nr": 3
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ar",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.6107152353968289,
+      "sentence_nr": 3
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ar",
@@ -17125,6 +18747,22 @@
       "score": 0.27307753334479423,
       "sentence_nr": 3
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ur",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.14965975078050625,
+      "sentence_nr": 3
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ur",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.22213502776474325,
+      "sentence_nr": 3
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ur",
@@ -17157,6 +18795,22 @@
       "score": 0.22213502776474325,
       "sentence_nr": 3
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ur",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.40842567407749947,
+      "sentence_nr": 3
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ur",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.49897421349092935,
+      "sentence_nr": 3
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ur",
@@ -17301,6 +18955,22 @@
       "score": 0.4510525482602028,
       "sentence_nr": 3
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "fr",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.21690365808279138,
+      "sentence_nr": 3
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "fr",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.33065826652002533,
+      "sentence_nr": 3
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "fr",
@@ -17333,6 +19003,22 @@
       "score": 0.27718461611705486,
       "sentence_nr": 3
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "fr",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.6230832293767097,
+      "sentence_nr": 3
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "fr",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.702540870003671,
+      "sentence_nr": 3
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "fr",
@@ -17477,6 +19163,22 @@
       "score": 0.42516173623967946,
       "sentence_nr": 3
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "bn",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.3685289119518548,
+      "sentence_nr": 3
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "bn",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.47062358557598893,
+      "sentence_nr": 3
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "bn",
@@ -17509,6 +19211,22 @@
       "score": 0.2249046365436241,
       "sentence_nr": 3
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "bn",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.4185938787651429,
+      "sentence_nr": 3
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "bn",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.6016367461945803,
+      "sentence_nr": 3
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "bn",
@@ -17653,6 +19371,22 @@
       "score": 0.32950116238735283,
       "sentence_nr": 3
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pt",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.20664181816537014,
+      "sentence_nr": 3
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pt",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.371359687688326,
+      "sentence_nr": 3
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "pt",
@@ -17685,6 +19419,22 @@
       "score": 0.5103516764863386,
       "sentence_nr": 3
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pt",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.6230832293767097,
+      "sentence_nr": 3
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pt",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.702540870003671,
+      "sentence_nr": 3
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "pt",
@@ -17829,6 +19579,22 @@
       "score": 0.0,
       "sentence_nr": 3
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pa",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 1.0,
+      "sentence_nr": 3
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pa",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 1.0,
+      "sentence_nr": 3
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "pa",
@@ -17861,6 +19627,22 @@
       "score": 1.0,
       "sentence_nr": 3
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pa",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 1.0,
+      "sentence_nr": 3
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pa",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 1.0,
+      "sentence_nr": 3
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "pa",
@@ -18005,6 +19787,22 @@
       "score": 0.3717184743596148,
       "sentence_nr": 3
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ru",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.22481074167380632,
+      "sentence_nr": 3
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ru",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.3761108267186685,
+      "sentence_nr": 3
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ru",
@@ -18037,6 +19835,22 @@
       "score": 0.2552663483401067,
       "sentence_nr": 3
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ru",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.3032929624979452,
+      "sentence_nr": 3
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ru",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.5077415447889289,
+      "sentence_nr": 3
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ru",
@@ -18181,6 +19995,22 @@
       "score": 0.27604929504751197,
       "sentence_nr": 3
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "sw",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.0,
+      "sentence_nr": 3
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "sw",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.007934677500708292,
+      "sentence_nr": 3
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "sw",
@@ -18213,6 +20043,22 @@
       "score": 0.2952752522340665,
       "sentence_nr": 3
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "sw",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.17729842264695017,
+      "sentence_nr": 3
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "sw",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.3337514618651578,
+      "sentence_nr": 3
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "sw",
@@ -18357,6 +20203,22 @@
       "score": 0.359573626731952,
       "sentence_nr": 3
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "id",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.13502367316243039,
+      "sentence_nr": 3
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "id",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.299859745321103,
+      "sentence_nr": 3
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "id",
@@ -18389,6 +20251,22 @@
       "score": 0.5667866238125795,
       "sentence_nr": 3
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "id",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.6230832293767097,
+      "sentence_nr": 3
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "id",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.702540870003671,
+      "sentence_nr": 3
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "id",
@@ -18533,6 +20411,22 @@
       "score": 0.301901669683193,
       "sentence_nr": 3
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "de",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.3166144686275811,
+      "sentence_nr": 3
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "de",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.4171828599209745,
+      "sentence_nr": 3
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "de",
@@ -18565,6 +20459,22 @@
       "score": 0.3284886849880412,
       "sentence_nr": 3
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "de",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.6230832293767097,
+      "sentence_nr": 3
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "de",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.702540870003671,
+      "sentence_nr": 3
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "de",
@@ -18709,6 +20619,22 @@
       "score": 0.2344095627038401,
       "sentence_nr": 3
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ja",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.0,
+      "sentence_nr": 3
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ja",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.248781805015534,
+      "sentence_nr": 3
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ja",
@@ -18741,6 +20667,22 @@
       "score": 0.21844360831325868,
       "sentence_nr": 3
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ja",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.4246183605185108,
+      "sentence_nr": 3
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ja",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.5497460511936695,
+      "sentence_nr": 3
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ja",
@@ -18885,6 +20827,22 @@
       "score": 0.2246029757863831,
       "sentence_nr": 4
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "en",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 1.0,
+      "sentence_nr": 4
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "en",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 1.0,
+      "sentence_nr": 4
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "en",
@@ -18917,6 +20875,22 @@
       "score": 0.9453478043428296,
       "sentence_nr": 4
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "en",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 1.0,
+      "sentence_nr": 4
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "en",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 1.0,
+      "sentence_nr": 4
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "en",
@@ -19061,6 +21035,22 @@
       "score": 0.5581982021478125,
       "sentence_nr": 4
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "zh",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.0,
+      "sentence_nr": 4
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "zh",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.6364887816884232,
+      "sentence_nr": 4
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "zh",
@@ -19093,6 +21083,22 @@
       "score": 0.629039349740581,
       "sentence_nr": 4
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "zh",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.0,
+      "sentence_nr": 4
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "zh",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.629039349740581,
+      "sentence_nr": 4
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "zh",
@@ -19237,6 +21243,22 @@
       "score": 0.37544324742239676,
       "sentence_nr": 4
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "hi",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.0,
+      "sentence_nr": 4
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "hi",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.5344225462130586,
+      "sentence_nr": 4
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "hi",
@@ -19269,6 +21291,22 @@
       "score": 0.41602211217571683,
       "sentence_nr": 4
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "hi",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.18842393723950338,
+      "sentence_nr": 4
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "hi",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.5646801080937621,
+      "sentence_nr": 4
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "hi",
@@ -19413,6 +21451,22 @@
       "score": 0.41477028165511615,
       "sentence_nr": 4
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "es",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.0,
+      "sentence_nr": 4
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "es",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.33319754264314433,
+      "sentence_nr": 4
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "es",
@@ -19445,6 +21499,22 @@
       "score": 0.4164061298971701,
       "sentence_nr": 4
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "es",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.0,
+      "sentence_nr": 4
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "es",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.40562290854898025,
+      "sentence_nr": 4
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "es",
@@ -19589,6 +21659,22 @@
       "score": 0.3644112480028862,
       "sentence_nr": 4
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ar",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.0,
+      "sentence_nr": 4
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ar",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.3128264071150837,
+      "sentence_nr": 4
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ar",
@@ -19621,6 +21707,22 @@
       "score": 0.393379300802006,
       "sentence_nr": 4
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ar",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.0,
+      "sentence_nr": 4
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ar",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.30325797018747325,
+      "sentence_nr": 4
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ar",
@@ -19765,6 +21867,22 @@
       "score": 0.511876122662448,
       "sentence_nr": 4
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ur",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.0,
+      "sentence_nr": 4
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ur",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.4418365362317144,
+      "sentence_nr": 4
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ur",
@@ -19797,6 +21915,22 @@
       "score": 0.5049375875723539,
       "sentence_nr": 4
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ur",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.0,
+      "sentence_nr": 4
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ur",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.511876122662448,
+      "sentence_nr": 4
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ur",
@@ -19941,6 +22075,22 @@
       "score": 0.3765697091436241,
       "sentence_nr": 4
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "fr",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.0,
+      "sentence_nr": 4
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "fr",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.3431841258656284,
+      "sentence_nr": 4
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "fr",
@@ -19973,6 +22123,22 @@
       "score": 0.6989238098201116,
       "sentence_nr": 4
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "fr",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.33118227522229554,
+      "sentence_nr": 4
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "fr",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.7010244056936935,
+      "sentence_nr": 4
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "fr",
@@ -20117,6 +22283,22 @@
       "score": 0.40052428191473877,
       "sentence_nr": 4
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "bn",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.0,
+      "sentence_nr": 4
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "bn",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.2630268050449769,
+      "sentence_nr": 4
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "bn",
@@ -20149,6 +22331,22 @@
       "score": 0.39336600752225864,
       "sentence_nr": 4
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "bn",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.0,
+      "sentence_nr": 4
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "bn",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.5189004396088754,
+      "sentence_nr": 4
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "bn",
@@ -20293,6 +22491,22 @@
       "score": 0.4746119151171374,
       "sentence_nr": 4
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pt",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.195647514979229,
+      "sentence_nr": 4
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pt",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.5832256253964303,
+      "sentence_nr": 4
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "pt",
@@ -20325,6 +22539,22 @@
       "score": 0.5639241776831634,
       "sentence_nr": 4
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pt",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.4513165758172509,
+      "sentence_nr": 4
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pt",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.7514771576902608,
+      "sentence_nr": 4
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "pt",
@@ -20469,6 +22699,22 @@
       "score": 0.4806367958084579,
       "sentence_nr": 4
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pa",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.0,
+      "sentence_nr": 4
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pa",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.5066311799500233,
+      "sentence_nr": 4
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "pa",
@@ -20501,6 +22747,22 @@
       "score": 0.5013632657267051,
       "sentence_nr": 4
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pa",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.0,
+      "sentence_nr": 4
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pa",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.49118110181149666,
+      "sentence_nr": 4
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "pa",
@@ -20645,6 +22907,22 @@
       "score": 0.40443357144012176,
       "sentence_nr": 4
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ru",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.0,
+      "sentence_nr": 4
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ru",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.4354398635855642,
+      "sentence_nr": 4
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ru",
@@ -20677,6 +22955,22 @@
       "score": 0.4121946181418776,
       "sentence_nr": 4
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ru",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.0,
+      "sentence_nr": 4
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ru",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.39909989628767284,
+      "sentence_nr": 4
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ru",
@@ -20821,6 +23115,22 @@
       "score": 0.514952316880994,
       "sentence_nr": 4
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "sw",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.21972813874997157,
+      "sentence_nr": 4
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "sw",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.6022094443409847,
+      "sentence_nr": 4
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "sw",
@@ -20853,6 +23163,22 @@
       "score": 0.3128496839849598,
       "sentence_nr": 4
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "sw",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.5773502691896258,
+      "sentence_nr": 4
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "sw",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.7999099314029202,
+      "sentence_nr": 4
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "sw",
@@ -20997,6 +23323,22 @@
       "score": 0.47465074831919213,
       "sentence_nr": 4
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "id",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.0,
+      "sentence_nr": 4
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "id",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.4425973012069069,
+      "sentence_nr": 4
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "id",
@@ -21029,6 +23371,22 @@
       "score": 0.4425973012069069,
       "sentence_nr": 4
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "id",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.0,
+      "sentence_nr": 4
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "id",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.4440931655950853,
+      "sentence_nr": 4
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "id",
@@ -21173,6 +23531,22 @@
       "score": 0.3830425592586042,
       "sentence_nr": 4
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "de",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.0,
+      "sentence_nr": 4
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "de",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.40508457369784023,
+      "sentence_nr": 4
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "de",
@@ -21205,6 +23579,22 @@
       "score": 0.3830425592586042,
       "sentence_nr": 4
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "de",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.0,
+      "sentence_nr": 4
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "de",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.3830425592586042,
+      "sentence_nr": 4
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "de",
@@ -21349,6 +23739,22 @@
       "score": 0.5521590062829653,
       "sentence_nr": 4
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ja",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.23693055763743093,
+      "sentence_nr": 4
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ja",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.6474126202050918,
+      "sentence_nr": 4
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ja",
@@ -21381,6 +23787,22 @@
       "score": 0.6131017059052001,
       "sentence_nr": 4
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ja",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.23693055763743093,
+      "sentence_nr": 4
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ja",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.6474126202050918,
+      "sentence_nr": 4
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ja",
@@ -21525,6 +23947,22 @@
       "score": 0.2381658499765768,
       "sentence_nr": 5
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "en",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 1.0,
+      "sentence_nr": 5
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "en",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 1.0,
+      "sentence_nr": 5
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "en",
@@ -21557,6 +23995,22 @@
       "score": 0.9063898435384111,
       "sentence_nr": 5
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "en",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 1.0,
+      "sentence_nr": 5
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "en",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 1.0,
+      "sentence_nr": 5
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "en",
@@ -21701,6 +24155,22 @@
       "score": 0.7538467008030766,
       "sentence_nr": 5
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "zh",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.47410002229034043,
+      "sentence_nr": 5
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "zh",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.7538467008030766,
+      "sentence_nr": 5
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "zh",
@@ -21733,6 +24203,22 @@
       "score": 0.7410180114887145,
       "sentence_nr": 5
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "zh",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.47375069012411286,
+      "sentence_nr": 5
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "zh",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.7543919667018285,
+      "sentence_nr": 5
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "zh",
@@ -21877,6 +24363,22 @@
       "score": 0.8597893117683423,
       "sentence_nr": 5
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "hi",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.4441961115027302,
+      "sentence_nr": 5
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "hi",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.7565542718609186,
+      "sentence_nr": 5
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "hi",
@@ -21909,6 +24411,22 @@
       "score": 0.7513336773729535,
       "sentence_nr": 5
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "hi",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.48181149445310956,
+      "sentence_nr": 5
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "hi",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.7675828789334244,
+      "sentence_nr": 5
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "hi",
@@ -22053,6 +24571,22 @@
       "score": 0.6681898017773897,
       "sentence_nr": 5
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "es",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.44897710722021167,
+      "sentence_nr": 5
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "es",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.6862249089515978,
+      "sentence_nr": 5
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "es",
@@ -22085,6 +24619,22 @@
       "score": 0.6392900613840917,
       "sentence_nr": 5
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "es",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.44897710722021167,
+      "sentence_nr": 5
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "es",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.6862249089515978,
+      "sentence_nr": 5
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "es",
@@ -22229,6 +24779,22 @@
       "score": 0.6207272323003366,
       "sentence_nr": 5
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ar",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.2704091953828695,
+      "sentence_nr": 5
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ar",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.6207272323003366,
+      "sentence_nr": 5
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ar",
@@ -22261,6 +24827,22 @@
       "score": 0.6207272323003366,
       "sentence_nr": 5
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ar",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.2704091953828695,
+      "sentence_nr": 5
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ar",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.6207272323003366,
+      "sentence_nr": 5
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ar",
@@ -22405,6 +24987,22 @@
       "score": 0.8597893117683423,
       "sentence_nr": 5
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ur",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.4727805712999679,
+      "sentence_nr": 5
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ur",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.7717158158167359,
+      "sentence_nr": 5
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ur",
@@ -22437,6 +25035,22 @@
       "score": 0.7689532399280165,
       "sentence_nr": 5
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ur",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.5465526716276092,
+      "sentence_nr": 5
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ur",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.8012679276648627,
+      "sentence_nr": 5
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ur",
@@ -22581,6 +25195,22 @@
       "score": 0.713787745993602,
       "sentence_nr": 5
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "fr",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.43007078231141604,
+      "sentence_nr": 5
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "fr",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.6680243275130087,
+      "sentence_nr": 5
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "fr",
@@ -22613,6 +25243,22 @@
       "score": 0.701102363286568,
       "sentence_nr": 5
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "fr",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.5896613549548209,
+      "sentence_nr": 5
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "fr",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.7528914749586836,
+      "sentence_nr": 5
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "fr",
@@ -22757,6 +25403,22 @@
       "score": 0.566236392445952,
       "sentence_nr": 5
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "bn",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.32965129549221617,
+      "sentence_nr": 5
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "bn",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.623436907204599,
+      "sentence_nr": 5
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "bn",
@@ -22789,6 +25451,22 @@
       "score": 0.623436907204599,
       "sentence_nr": 5
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "bn",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.32078739729528816,
+      "sentence_nr": 5
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "bn",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.5817366082116868,
+      "sentence_nr": 5
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "bn",
@@ -22933,6 +25611,22 @@
       "score": 0.5788026000794341,
       "sentence_nr": 5
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pt",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.3665134361137304,
+      "sentence_nr": 5
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pt",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.6118771029352303,
+      "sentence_nr": 5
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "pt",
@@ -22965,6 +25659,22 @@
       "score": 0.6090575371936678,
       "sentence_nr": 5
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pt",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.4085639059221913,
+      "sentence_nr": 5
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pt",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.5887623870312143,
+      "sentence_nr": 5
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "pt",
@@ -23109,6 +25819,22 @@
       "score": 0.7469480084357536,
       "sentence_nr": 5
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pa",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.42612283570374254,
+      "sentence_nr": 5
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pa",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.7185121839177114,
+      "sentence_nr": 5
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "pa",
@@ -23141,6 +25867,22 @@
       "score": 0.7133166401137868,
       "sentence_nr": 5
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pa",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.6917901740466924,
+      "sentence_nr": 5
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pa",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.8479928839177578,
+      "sentence_nr": 5
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "pa",
@@ -23285,6 +26027,22 @@
       "score": 0.5609439249510223,
       "sentence_nr": 5
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ru",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.21690365808279138,
+      "sentence_nr": 5
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ru",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.5502221839528678,
+      "sentence_nr": 5
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ru",
@@ -23317,6 +26075,22 @@
       "score": 0.540043957078071,
       "sentence_nr": 5
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ru",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.3282518529729176,
+      "sentence_nr": 5
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ru",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.6453010665294326,
+      "sentence_nr": 5
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ru",
@@ -23461,6 +26235,22 @@
       "score": 0.6578570934289981,
       "sentence_nr": 5
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "sw",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.24456656109396324,
+      "sentence_nr": 5
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "sw",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.629934465484704,
+      "sentence_nr": 5
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "sw",
@@ -23493,6 +26283,22 @@
       "score": 0.6570214418399444,
       "sentence_nr": 5
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "sw",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.6888365053466561,
+      "sentence_nr": 5
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "sw",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.8656273480576243,
+      "sentence_nr": 5
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "sw",
@@ -23637,6 +26443,22 @@
       "score": 0.7516103467926585,
       "sentence_nr": 5
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "id",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.45307778036928104,
+      "sentence_nr": 5
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "id",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.6935397252637394,
+      "sentence_nr": 5
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "id",
@@ -23669,6 +26491,22 @@
       "score": 0.7516103467926585,
       "sentence_nr": 5
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "id",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.8492326635760689,
+      "sentence_nr": 5
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "id",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.9027320255916917,
+      "sentence_nr": 5
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "id",
@@ -23813,6 +26651,22 @@
       "score": 0.7361065921505279,
       "sentence_nr": 5
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "de",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.5072784644062104,
+      "sentence_nr": 5
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "de",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.7361065921505279,
+      "sentence_nr": 5
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "de",
@@ -23845,6 +26699,22 @@
       "score": 0.7361065921505279,
       "sentence_nr": 5
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "de",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.7196315267102845,
+      "sentence_nr": 5
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "de",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.8835331636515565,
+      "sentence_nr": 5
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "de",
@@ -23989,6 +26859,22 @@
       "score": 0.7062510642584722,
       "sentence_nr": 5
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ja",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.1624355752882384,
+      "sentence_nr": 5
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ja",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.5090396683756193,
+      "sentence_nr": 5
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ja",
@@ -24021,6 +26907,22 @@
       "score": 0.6470050797908481,
       "sentence_nr": 5
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ja",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.45506803308128024,
+      "sentence_nr": 5
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ja",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.7117510256855165,
+      "sentence_nr": 5
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ja",
@@ -24165,6 +27067,22 @@
       "score": 0.3263040636562357,
       "sentence_nr": 6
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "en",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.7511573912724299,
+      "sentence_nr": 6
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "en",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.9453473543978153,
+      "sentence_nr": 6
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "en",
@@ -24197,6 +27115,22 @@
       "score": 0.9453473543978153,
       "sentence_nr": 6
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "en",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.7511573912724299,
+      "sentence_nr": 6
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "en",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.9453473543978153,
+      "sentence_nr": 6
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "en",
@@ -24341,6 +27275,22 @@
       "score": 0.551397074868541,
       "sentence_nr": 6
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "zh",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.19464521962073492,
+      "sentence_nr": 6
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "zh",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.5838790966762375,
+      "sentence_nr": 6
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "zh",
@@ -24373,6 +27323,22 @@
       "score": 0.5403400891349619,
       "sentence_nr": 6
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "zh",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.3060368950930089,
+      "sentence_nr": 6
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "zh",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.6498981440676681,
+      "sentence_nr": 6
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "zh",
@@ -24517,6 +27483,22 @@
       "score": 0.6589376390020449,
       "sentence_nr": 6
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "hi",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.43021236941942204,
+      "sentence_nr": 6
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "hi",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.7142896582178452,
+      "sentence_nr": 6
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "hi",
@@ -24549,6 +27531,22 @@
       "score": 0.6589376390020449,
       "sentence_nr": 6
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "hi",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.6255340042200862,
+      "sentence_nr": 6
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "hi",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.8724783049357475,
+      "sentence_nr": 6
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "hi",
@@ -24693,6 +27691,22 @@
       "score": 0.6509517796070665,
       "sentence_nr": 6
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "es",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.4481489512240194,
+      "sentence_nr": 6
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "es",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.7745649676018984,
+      "sentence_nr": 6
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "es",
@@ -24725,6 +27739,22 @@
       "score": 0.6509517796070665,
       "sentence_nr": 6
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "es",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.32365795029773287,
+      "sentence_nr": 6
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "es",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.6509517796070665,
+      "sentence_nr": 6
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "es",
@@ -24869,6 +27899,22 @@
       "score": 0.7968980206907678,
       "sentence_nr": 6
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ar",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.44787223195695314,
+      "sentence_nr": 6
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ar",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.7968980206907678,
+      "sentence_nr": 6
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ar",
@@ -24901,6 +27947,22 @@
       "score": 0.7968980206907678,
       "sentence_nr": 6
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ar",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.44787223195695314,
+      "sentence_nr": 6
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ar",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.7968980206907678,
+      "sentence_nr": 6
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ar",
@@ -25045,6 +28107,22 @@
       "score": 0.4800889669735933,
       "sentence_nr": 6
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ur",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.3347189874003768,
+      "sentence_nr": 6
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ur",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.6644203374869264,
+      "sentence_nr": 6
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ur",
@@ -25077,6 +28155,22 @@
       "score": 0.584996891148118,
       "sentence_nr": 6
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ur",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.5383680940297331,
+      "sentence_nr": 6
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ur",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.786096406361039,
+      "sentence_nr": 6
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ur",
@@ -25221,6 +28315,22 @@
       "score": 0.630711601223299,
       "sentence_nr": 6
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "fr",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.23198210427894825,
+      "sentence_nr": 6
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "fr",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.630711601223299,
+      "sentence_nr": 6
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "fr",
@@ -25253,6 +28363,22 @@
       "score": 0.630711601223299,
       "sentence_nr": 6
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "fr",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.23198210427894825,
+      "sentence_nr": 6
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "fr",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.630711601223299,
+      "sentence_nr": 6
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "fr",
@@ -25397,6 +28523,22 @@
       "score": 0.6985308026285912,
       "sentence_nr": 6
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "bn",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.42984824697674956,
+      "sentence_nr": 6
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "bn",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.7369844404912368,
+      "sentence_nr": 6
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "bn",
@@ -25429,6 +28571,22 @@
       "score": 0.7369844404912368,
       "sentence_nr": 6
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "bn",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.595092211343687,
+      "sentence_nr": 6
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "bn",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.7945212279546889,
+      "sentence_nr": 6
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "bn",
@@ -25573,6 +28731,22 @@
       "score": 0.6590438071804039,
       "sentence_nr": 6
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pt",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.44787223195695314,
+      "sentence_nr": 6
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pt",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.7968980206907678,
+      "sentence_nr": 6
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "pt",
@@ -25605,6 +28779,22 @@
       "score": 0.7968980206907678,
       "sentence_nr": 6
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pt",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.44787223195695314,
+      "sentence_nr": 6
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pt",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.7968980206907678,
+      "sentence_nr": 6
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "pt",
@@ -25749,6 +28939,22 @@
       "score": 0.5397693417183738,
       "sentence_nr": 6
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pa",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.421151249507493,
+      "sentence_nr": 6
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pa",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.6889522290200047,
+      "sentence_nr": 6
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "pa",
@@ -25781,6 +28987,22 @@
       "score": 0.6674242019044293,
       "sentence_nr": 6
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pa",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.5383680940297331,
+      "sentence_nr": 6
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pa",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.786096406361039,
+      "sentence_nr": 6
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "pa",
@@ -25925,6 +29147,22 @@
       "score": 0.58198979036704,
       "sentence_nr": 6
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ru",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.24047860794644352,
+      "sentence_nr": 6
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ru",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.58198979036704,
+      "sentence_nr": 6
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ru",
@@ -25957,6 +29195,22 @@
       "score": 0.58198979036704,
       "sentence_nr": 6
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ru",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.24047860794644352,
+      "sentence_nr": 6
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ru",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.58198979036704,
+      "sentence_nr": 6
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ru",
@@ -26101,6 +29355,22 @@
       "score": 0.598503332887995,
       "sentence_nr": 6
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "sw",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.3942058093215873,
+      "sentence_nr": 6
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "sw",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.6619197609807324,
+      "sentence_nr": 6
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "sw",
@@ -26133,6 +29403,22 @@
       "score": 0.6085546680624175,
       "sentence_nr": 6
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "sw",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.6173766800527999,
+      "sentence_nr": 6
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "sw",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.857390040146912,
+      "sentence_nr": 6
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "sw",
@@ -26277,6 +29563,22 @@
       "score": 0.5582775802710993,
       "sentence_nr": 6
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "id",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.0,
+      "sentence_nr": 6
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "id",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.5582775802710993,
+      "sentence_nr": 6
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "id",
@@ -26309,6 +29611,22 @@
       "score": 0.5582775802710993,
       "sentence_nr": 6
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "id",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.22436571657855092,
+      "sentence_nr": 6
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "id",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.61166969974579,
+      "sentence_nr": 6
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "id",
@@ -26453,6 +29771,22 @@
       "score": 0.8215788698315908,
       "sentence_nr": 6
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "de",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.5967384019266717,
+      "sentence_nr": 6
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "de",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.8544348080833218,
+      "sentence_nr": 6
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "de",
@@ -26485,6 +29819,22 @@
       "score": 0.7317734491561229,
       "sentence_nr": 6
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "de",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.4881010344921759,
+      "sentence_nr": 6
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "de",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.7317734491561229,
+      "sentence_nr": 6
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "de",
@@ -26629,6 +29979,22 @@
       "score": 0.6498499527552988,
       "sentence_nr": 6
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ja",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.29170205300854224,
+      "sentence_nr": 6
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ja",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.6498499527552988,
+      "sentence_nr": 6
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ja",
@@ -26661,6 +30027,22 @@
       "score": 0.6782734900436637,
       "sentence_nr": 6
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ja",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.4284945090100314,
+      "sentence_nr": 6
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ja",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.7246227738353674,
+      "sentence_nr": 6
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ja",
@@ -26805,6 +30187,22 @@
       "score": 0.33602633953270183,
       "sentence_nr": 7
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "en",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.10316589800033629,
+      "sentence_nr": 7
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "en",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.3117209570799365,
+      "sentence_nr": 7
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "en",
@@ -26837,6 +30235,22 @@
       "score": 0.03037224815656603,
       "sentence_nr": 7
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "en",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.306354798881405,
+      "sentence_nr": 7
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "en",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.5275640293284548,
+      "sentence_nr": 7
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "en",
@@ -26981,6 +30395,22 @@
       "score": 0.43272151570555034,
       "sentence_nr": 7
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "zh",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.09632940954518097,
+      "sentence_nr": 7
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "zh",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.3503787449131298,
+      "sentence_nr": 7
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "zh",
@@ -27013,6 +30443,22 @@
       "score": 0.05918530850500025,
       "sentence_nr": 7
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "zh",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.22815217447604735,
+      "sentence_nr": 7
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "zh",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.48973384754323573,
+      "sentence_nr": 7
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "zh",
@@ -27157,6 +30603,22 @@
       "score": 0.3458120002305796,
       "sentence_nr": 7
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "hi",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.05109780779294313,
+      "sentence_nr": 7
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "hi",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.27453721503960304,
+      "sentence_nr": 7
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "hi",
@@ -27189,6 +30651,22 @@
       "score": 0.20031726728306523,
       "sentence_nr": 7
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "hi",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.28020858443704566,
+      "sentence_nr": 7
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "hi",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.4608753441128863,
+      "sentence_nr": 7
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "hi",
@@ -27333,6 +30811,22 @@
       "score": 0.3364703638684802,
       "sentence_nr": 7
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "es",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.14974627745213473,
+      "sentence_nr": 7
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "es",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.3288440080133683,
+      "sentence_nr": 7
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "es",
@@ -27365,6 +30859,22 @@
       "score": 0.28912109037408523,
       "sentence_nr": 7
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "es",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.22274170049761707,
+      "sentence_nr": 7
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "es",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.4532692581443855,
+      "sentence_nr": 7
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "es",
@@ -27509,6 +31019,22 @@
       "score": 0.4017968725013381,
       "sentence_nr": 7
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ar",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.13122945980689196,
+      "sentence_nr": 7
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ar",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.3219448107305951,
+      "sentence_nr": 7
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ar",
@@ -27541,6 +31067,22 @@
       "score": 0.04151505758906764,
       "sentence_nr": 7
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ar",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.257476399230248,
+      "sentence_nr": 7
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ar",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.4616305811028715,
+      "sentence_nr": 7
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ar",
@@ -27685,6 +31227,22 @@
       "score": 0.3422914837190449,
       "sentence_nr": 7
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ur",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.0985948810698831,
+      "sentence_nr": 7
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ur",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.2593923911428194,
+      "sentence_nr": 7
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ur",
@@ -27717,6 +31275,22 @@
       "score": 0.2874483621307283,
       "sentence_nr": 7
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ur",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.28562106983744195,
+      "sentence_nr": 7
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ur",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.4749518446683913,
+      "sentence_nr": 7
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ur",
@@ -27861,6 +31435,22 @@
       "score": 0.3121110160693956,
       "sentence_nr": 7
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "fr",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.05658523178050362,
+      "sentence_nr": 7
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "fr",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.27543168466680934,
+      "sentence_nr": 7
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "fr",
@@ -27893,6 +31483,22 @@
       "score": 0.011560595536104562,
       "sentence_nr": 7
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "fr",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.28619616266290565,
+      "sentence_nr": 7
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "fr",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.509863149159976,
+      "sentence_nr": 7
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "fr",
@@ -28037,6 +31643,22 @@
       "score": 0.3791567776918788,
       "sentence_nr": 7
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "bn",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.111871161288306,
+      "sentence_nr": 7
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "bn",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.30886933073849715,
+      "sentence_nr": 7
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "bn",
@@ -28069,6 +31691,22 @@
       "score": 0.09453698369211004,
       "sentence_nr": 7
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "bn",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.266802610525441,
+      "sentence_nr": 7
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "bn",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.4519949729282707,
+      "sentence_nr": 7
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "bn",
@@ -28213,6 +31851,22 @@
       "score": 0.31793222329793575,
       "sentence_nr": 7
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pt",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.008104773738922768,
+      "sentence_nr": 7
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pt",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.07651300648420487,
+      "sentence_nr": 7
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "pt",
@@ -28245,6 +31899,22 @@
       "score": 0.27004759126600675,
       "sentence_nr": 7
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pt",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.18740401341623053,
+      "sentence_nr": 7
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pt",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.47595080320432986,
+      "sentence_nr": 7
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "pt",
@@ -28389,6 +32059,22 @@
       "score": 0.3846197304420823,
       "sentence_nr": 7
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pa",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.10793175369072293,
+      "sentence_nr": 7
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pa",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.29660617805108785,
+      "sentence_nr": 7
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "pa",
@@ -28421,6 +32107,22 @@
       "score": 0.07485928007606017,
       "sentence_nr": 7
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pa",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.26242764207671093,
+      "sentence_nr": 7
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pa",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.48831610046795776,
+      "sentence_nr": 7
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "pa",
@@ -28565,6 +32267,22 @@
       "score": 0.3902085179927465,
       "sentence_nr": 7
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ru",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.12351824822447698,
+      "sentence_nr": 7
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ru",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.3906003941586503,
+      "sentence_nr": 7
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ru",
@@ -28597,6 +32315,22 @@
       "score": 0.21925629669878902,
       "sentence_nr": 7
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ru",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.2730720849895809,
+      "sentence_nr": 7
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ru",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.4904647237477465,
+      "sentence_nr": 7
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ru",
@@ -28741,6 +32475,22 @@
       "score": 0.34378295878971765,
       "sentence_nr": 7
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "sw",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.04731666745775026,
+      "sentence_nr": 7
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "sw",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.263716605484278,
+      "sentence_nr": 7
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "sw",
@@ -28773,6 +32523,22 @@
       "score": 0.2374647159547877,
       "sentence_nr": 7
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "sw",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.21356649685679252,
+      "sentence_nr": 7
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "sw",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.4243725761948095,
+      "sentence_nr": 7
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "sw",
@@ -28917,6 +32683,22 @@
       "score": 0.4035796398628449,
       "sentence_nr": 7
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "id",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.13346878790838718,
+      "sentence_nr": 7
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "id",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.30809134673210076,
+      "sentence_nr": 7
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "id",
@@ -28949,6 +32731,22 @@
       "score": 0.09384599631616997,
       "sentence_nr": 7
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "id",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.26356108382670784,
+      "sentence_nr": 7
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "id",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.47640029725892175,
+      "sentence_nr": 7
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "id",
@@ -29093,6 +32891,22 @@
       "score": 0.356756117753337,
       "sentence_nr": 7
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "de",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.059890467169145326,
+      "sentence_nr": 7
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "de",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.2852233544262429,
+      "sentence_nr": 7
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "de",
@@ -29125,6 +32939,22 @@
       "score": 0.05915285533036862,
       "sentence_nr": 7
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "de",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.2529913062741307,
+      "sentence_nr": 7
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "de",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.5047314299093542,
+      "sentence_nr": 7
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "de",
@@ -29269,6 +33099,22 @@
       "score": 0.36682227371085463,
       "sentence_nr": 7
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ja",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.11509582662256548,
+      "sentence_nr": 7
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ja",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.2567415288537121,
+      "sentence_nr": 7
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ja",
@@ -29301,6 +33147,22 @@
       "score": 0.2495519218392036,
       "sentence_nr": 7
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ja",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.2700569573710134,
+      "sentence_nr": 7
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ja",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.4456029881233736,
+      "sentence_nr": 7
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ja",
@@ -29445,6 +33307,22 @@
       "score": 0.4142901090120915,
       "sentence_nr": 8
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "en",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.11719733243527225,
+      "sentence_nr": 8
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "en",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.4023449434976014,
+      "sentence_nr": 8
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "en",
@@ -29477,6 +33355,22 @@
       "score": 0.32069132319909655,
       "sentence_nr": 8
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "en",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.47039611627724026,
+      "sentence_nr": 8
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "en",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.6298835480734482,
+      "sentence_nr": 8
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "en",
@@ -29621,6 +33515,22 @@
       "score": 0.41085414309816914,
       "sentence_nr": 8
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "zh",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.0,
+      "sentence_nr": 8
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "zh",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.24522672479887386,
+      "sentence_nr": 8
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "zh",
@@ -29653,6 +33563,22 @@
       "score": 0.4323734152924571,
       "sentence_nr": 8
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "zh",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.44212827861876997,
+      "sentence_nr": 8
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "zh",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.5843055590705747,
+      "sentence_nr": 8
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "zh",
@@ -29797,6 +33723,22 @@
       "score": 0.5244854229988815,
       "sentence_nr": 8
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "hi",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.30890092021323623,
+      "sentence_nr": 8
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "hi",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.5676965183365866,
+      "sentence_nr": 8
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "hi",
@@ -29829,6 +33771,22 @@
       "score": 0.6213537794704693,
       "sentence_nr": 8
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "hi",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.43186481103649477,
+      "sentence_nr": 8
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "hi",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.5834549494301647,
+      "sentence_nr": 8
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "hi",
@@ -29973,6 +33931,22 @@
       "score": 0.37017501464955627,
       "sentence_nr": 8
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "es",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.23874935220945062,
+      "sentence_nr": 8
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "es",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.4641140866580373,
+      "sentence_nr": 8
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "es",
@@ -30005,6 +33979,22 @@
       "score": 0.5031092445628172,
       "sentence_nr": 8
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "es",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.4094746713677566,
+      "sentence_nr": 8
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "es",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.5831913487431972,
+      "sentence_nr": 8
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "es",
@@ -30149,6 +34139,22 @@
       "score": 0.3750035199199742,
       "sentence_nr": 8
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ar",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.2305720744624697,
+      "sentence_nr": 8
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ar",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.4655848022427904,
+      "sentence_nr": 8
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ar",
@@ -30181,6 +34187,22 @@
       "score": 0.49618971681248764,
       "sentence_nr": 8
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ar",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.45479124441660884,
+      "sentence_nr": 8
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ar",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.5872444902217987,
+      "sentence_nr": 8
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ar",
@@ -30325,6 +34347,22 @@
       "score": 0.44334313717706003,
       "sentence_nr": 8
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ur",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.20498415630763028,
+      "sentence_nr": 8
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ur",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.41559166314823337,
+      "sentence_nr": 8
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ur",
@@ -30357,6 +34395,22 @@
       "score": 0.3253161209971999,
       "sentence_nr": 8
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ur",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.533167536340577,
+      "sentence_nr": 8
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ur",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.6607380945724757,
+      "sentence_nr": 8
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ur",
@@ -30501,6 +34555,22 @@
       "score": 0.5439625482235064,
       "sentence_nr": 8
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "fr",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.40066361284662694,
+      "sentence_nr": 8
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "fr",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.5020331850026643,
+      "sentence_nr": 8
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "fr",
@@ -30533,6 +34603,22 @@
       "score": 0.5043062352893725,
       "sentence_nr": 8
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "fr",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.36330554542576643,
+      "sentence_nr": 8
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "fr",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.5680915937757648,
+      "sentence_nr": 8
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "fr",
@@ -30677,6 +34763,22 @@
       "score": 0.29886658673327365,
       "sentence_nr": 8
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "bn",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.0,
+      "sentence_nr": 8
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "bn",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.18485608959451436,
+      "sentence_nr": 8
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "bn",
@@ -30709,6 +34811,22 @@
       "score": 0.3943841419148219,
       "sentence_nr": 8
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "bn",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.3890858187033785,
+      "sentence_nr": 8
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "bn",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.517430128986546,
+      "sentence_nr": 8
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "bn",
@@ -30853,6 +34971,22 @@
       "score": 0.5044329486461447,
       "sentence_nr": 8
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pt",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.3026681380066168,
+      "sentence_nr": 8
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pt",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.492659495510912,
+      "sentence_nr": 8
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "pt",
@@ -30885,6 +35019,22 @@
       "score": 0.44142087654422146,
       "sentence_nr": 8
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pt",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.43218646131638366,
+      "sentence_nr": 8
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pt",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.5402913319043152,
+      "sentence_nr": 8
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "pt",
@@ -31029,6 +35179,22 @@
       "score": 0.0067104198717751464,
       "sentence_nr": 8
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pa",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 1.0,
+      "sentence_nr": 8
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pa",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 1.0,
+      "sentence_nr": 8
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "pa",
@@ -31061,6 +35227,22 @@
       "score": 0.9169897590736298,
       "sentence_nr": 8
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pa",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 1.0,
+      "sentence_nr": 8
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pa",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 1.0,
+      "sentence_nr": 8
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "pa",
@@ -31205,6 +35387,22 @@
       "score": 0.2628849077177109,
       "sentence_nr": 8
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ru",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.34613789243685805,
+      "sentence_nr": 8
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ru",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.49330323192607783,
+      "sentence_nr": 8
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ru",
@@ -31237,6 +35435,22 @@
       "score": 0.2568045428196672,
       "sentence_nr": 8
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ru",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.43218646131638366,
+      "sentence_nr": 8
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ru",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.5369715393520321,
+      "sentence_nr": 8
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ru",
@@ -31381,6 +35595,22 @@
       "score": 0.16448947606185552,
       "sentence_nr": 8
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "sw",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.30630098078522544,
+      "sentence_nr": 8
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "sw",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.506196410096354,
+      "sentence_nr": 8
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "sw",
@@ -31413,6 +35643,22 @@
       "score": 0.24615921057796505,
       "sentence_nr": 8
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "sw",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.4955056754877292,
+      "sentence_nr": 8
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "sw",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.6304896503844739,
+      "sentence_nr": 8
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "sw",
@@ -31557,6 +35803,22 @@
       "score": 0.190140357671548,
       "sentence_nr": 8
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "id",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.16776974914462364,
+      "sentence_nr": 8
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "id",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.39918336227233053,
+      "sentence_nr": 8
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "id",
@@ -31589,6 +35851,22 @@
       "score": 0.41650969469918997,
       "sentence_nr": 8
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "id",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.4067965564160296,
+      "sentence_nr": 8
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "id",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.5347497635970215,
+      "sentence_nr": 8
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "id",
@@ -31733,6 +36011,22 @@
       "score": 0.49625515445592083,
       "sentence_nr": 8
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "de",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.33626819961829335,
+      "sentence_nr": 8
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "de",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.5531226519754557,
+      "sentence_nr": 8
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "de",
@@ -31765,6 +36059,22 @@
       "score": 0.4437597552815582,
       "sentence_nr": 8
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "de",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.5004894768464765,
+      "sentence_nr": 8
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "de",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.6103735933609515,
+      "sentence_nr": 8
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "de",
@@ -31909,6 +36219,22 @@
       "score": 0.4186167762559285,
       "sentence_nr": 8
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ja",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.3228858965814099,
+      "sentence_nr": 8
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ja",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.4495802766763041,
+      "sentence_nr": 8
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ja",
@@ -31941,6 +36267,22 @@
       "score": 0.3796021685415706,
       "sentence_nr": 8
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ja",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.4317853842116786,
+      "sentence_nr": 8
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ja",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.5396321094501078,
+      "sentence_nr": 8
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ja",
@@ -32085,6 +36427,22 @@
       "score": 0.013538497707846785,
       "sentence_nr": 9
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "en",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.3314824344065458,
+      "sentence_nr": 9
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "en",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.47145091131338446,
+      "sentence_nr": 9
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "en",
@@ -32117,6 +36475,22 @@
       "score": 0.38156158663679846,
       "sentence_nr": 9
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "en",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.26279137685269766,
+      "sentence_nr": 9
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "en",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.4760220740362435,
+      "sentence_nr": 9
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "en",
@@ -32261,6 +36635,22 @@
       "score": 0.08163977068875294,
       "sentence_nr": 9
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "zh",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.17431417316164047,
+      "sentence_nr": 9
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "zh",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.2946006716848339,
+      "sentence_nr": 9
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "zh",
@@ -32293,6 +36683,22 @@
       "score": 0.22948919855739472,
       "sentence_nr": 9
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "zh",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.17600429416656618,
+      "sentence_nr": 9
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "zh",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.2712560798864272,
+      "sentence_nr": 9
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "zh",
@@ -32437,6 +36843,22 @@
       "score": 0.3156355830822428,
       "sentence_nr": 9
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "hi",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.15110567441923345,
+      "sentence_nr": 9
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "hi",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.3452827306773606,
+      "sentence_nr": 9
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "hi",
@@ -32469,6 +36891,22 @@
       "score": 0.3818322535970043,
       "sentence_nr": 9
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "hi",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.16091123830242154,
+      "sentence_nr": 9
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "hi",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.3479090205359599,
+      "sentence_nr": 9
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "hi",
@@ -32613,6 +37051,22 @@
       "score": 0.022303919896869945,
       "sentence_nr": 9
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "es",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.12503614625842938,
+      "sentence_nr": 9
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "es",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.3600940511104839,
+      "sentence_nr": 9
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "es",
@@ -32645,6 +37099,22 @@
       "score": 0.31758120882708796,
       "sentence_nr": 9
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "es",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.0,
+      "sentence_nr": 9
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "es",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.31669205297743036,
+      "sentence_nr": 9
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "es",
@@ -32789,6 +37259,22 @@
       "score": 0.013915288440632284,
       "sentence_nr": 9
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ar",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.27011759273645686,
+      "sentence_nr": 9
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ar",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.4427599081001661,
+      "sentence_nr": 9
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ar",
@@ -32821,6 +37307,22 @@
       "score": 0.39112369376374106,
       "sentence_nr": 9
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ar",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.25299682930744943,
+      "sentence_nr": 9
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ar",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.4798320133488269,
+      "sentence_nr": 9
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ar",
@@ -32965,6 +37467,22 @@
       "score": 0.01536966738773372,
       "sentence_nr": 9
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ur",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.0,
+      "sentence_nr": 9
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ur",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.2808927299109351,
+      "sentence_nr": 9
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ur",
@@ -32997,6 +37515,22 @@
       "score": 0.32059338352121075,
       "sentence_nr": 9
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ur",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.24951692246125404,
+      "sentence_nr": 9
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ur",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.5427942390000431,
+      "sentence_nr": 9
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ur",
@@ -33141,6 +37675,22 @@
       "score": 0.007281906895508523,
       "sentence_nr": 9
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "fr",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.26632240818317526,
+      "sentence_nr": 9
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "fr",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.467076790922237,
+      "sentence_nr": 9
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "fr",
@@ -33173,6 +37723,22 @@
       "score": 0.4340281226634826,
       "sentence_nr": 9
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "fr",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.0,
+      "sentence_nr": 9
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "fr",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.3857801012055859,
+      "sentence_nr": 9
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "fr",
@@ -33317,6 +37883,22 @@
       "score": 0.01357525601063516,
       "sentence_nr": 9
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "bn",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.25387990321843446,
+      "sentence_nr": 9
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "bn",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.4061799423946215,
+      "sentence_nr": 9
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "bn",
@@ -33349,6 +37931,22 @@
       "score": 0.382987159925022,
       "sentence_nr": 9
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "bn",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.0,
+      "sentence_nr": 9
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "bn",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.36557640203818875,
+      "sentence_nr": 9
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "bn",
@@ -33493,6 +38091,22 @@
       "score": 0.022303919896869945,
       "sentence_nr": 9
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pt",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.3882967156327901,
+      "sentence_nr": 9
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pt",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.4287492504761661,
+      "sentence_nr": 9
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "pt",
@@ -33525,6 +38139,22 @@
       "score": 0.2756316951639811,
       "sentence_nr": 9
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pt",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.0,
+      "sentence_nr": 9
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pt",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.2778060655126336,
+      "sentence_nr": 9
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "pt",
@@ -33669,6 +38299,22 @@
       "score": 0.18039960295364865,
       "sentence_nr": 9
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pa",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.13602652550459576,
+      "sentence_nr": 9
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pa",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.3002035243994678,
+      "sentence_nr": 9
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "pa",
@@ -33701,6 +38347,22 @@
       "score": 0.3078571099929154,
       "sentence_nr": 9
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pa",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.18061023425907288,
+      "sentence_nr": 9
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pa",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.3684591961152223,
+      "sentence_nr": 9
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "pa",
@@ -33845,6 +38507,22 @@
       "score": 0.01943377856541192,
       "sentence_nr": 9
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ru",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.24250789663911215,
+      "sentence_nr": 9
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ru",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.44127658727390434,
+      "sentence_nr": 9
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ru",
@@ -33877,6 +38555,22 @@
       "score": 0.4690517750319636,
       "sentence_nr": 9
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ru",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.1743771229292808,
+      "sentence_nr": 9
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ru",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.46893502773403367,
+      "sentence_nr": 9
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ru",
@@ -34021,6 +38715,22 @@
       "score": 0.014749122939855126,
       "sentence_nr": 9
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "sw",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.0,
+      "sentence_nr": 9
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "sw",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.28355113133330917,
+      "sentence_nr": 9
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "sw",
@@ -34053,6 +38763,22 @@
       "score": 0.22585782564798598,
       "sentence_nr": 9
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "sw",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.0,
+      "sentence_nr": 9
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "sw",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.2753348107620908,
+      "sentence_nr": 9
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "sw",
@@ -34197,6 +38923,22 @@
       "score": 0.022303919896869945,
       "sentence_nr": 9
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "id",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.18723860296151318,
+      "sentence_nr": 9
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "id",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.3744699636940152,
+      "sentence_nr": 9
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "id",
@@ -34229,6 +38971,22 @@
       "score": 0.3295167855876769,
       "sentence_nr": 9
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "id",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.25753379048739855,
+      "sentence_nr": 9
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "id",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.42531177875262893,
+      "sentence_nr": 9
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "id",
@@ -34373,6 +39131,22 @@
       "score": 0.013501937941345124,
       "sentence_nr": 9
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "de",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.12832055613623328,
+      "sentence_nr": 9
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "de",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.239266757938952,
+      "sentence_nr": 9
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "de",
@@ -34405,6 +39179,22 @@
       "score": 0.3791177761741048,
       "sentence_nr": 9
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "de",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.0,
+      "sentence_nr": 9
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "de",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.3767372261720185,
+      "sentence_nr": 9
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "de",
@@ -34549,6 +39339,22 @@
       "score": 0.8142101616656354,
       "sentence_nr": 9
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ja",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 0.0,
+      "sentence_nr": 9
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ja",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 0.2762822897608569,
+      "sentence_nr": 9
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ja",
@@ -34581,6 +39387,22 @@
       "score": 1.0,
       "sentence_nr": 9
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ja",
+      "task": "translation_from",
+      "metric": "bleu",
+      "score": 1.0,
+      "sentence_nr": 9
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ja",
+      "task": "translation_from",
+      "metric": "chrf",
+      "score": 1.0,
+      "sentence_nr": 9
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ja",
@@ -34725,6 +39547,22 @@
       "score": 0.77238965036654,
       "sentence_nr": 0
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "en",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.48244543610473856,
+      "sentence_nr": 0
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "en",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.7294391805717774,
+      "sentence_nr": 0
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "en",
@@ -34757,6 +39595,22 @@
       "score": 0.7990339788905771,
       "sentence_nr": 0
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "en",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.5938611220262585,
+      "sentence_nr": 0
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "en",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.8079474861665713,
+      "sentence_nr": 0
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "en",
@@ -34901,6 +39755,22 @@
       "score": 0.3411361400094189,
       "sentence_nr": 0
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "zh",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.3504650671187503,
+      "sentence_nr": 0
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "zh",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.3985617531166433,
+      "sentence_nr": 0
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "zh",
@@ -34933,6 +39803,22 @@
       "score": 0.3355531727847081,
       "sentence_nr": 0
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "zh",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.4317270147628918,
+      "sentence_nr": 0
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "zh",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.4476891051893246,
+      "sentence_nr": 0
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "zh",
@@ -35077,6 +39963,22 @@
       "score": 0.9463396364218181,
       "sentence_nr": 0
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "hi",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.8562379115188704,
+      "sentence_nr": 0
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "hi",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.914880147320643,
+      "sentence_nr": 0
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "hi",
@@ -35109,6 +40011,22 @@
       "score": 0.9958930217841712,
       "sentence_nr": 0
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "hi",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.9878765474230741,
+      "sentence_nr": 0
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "hi",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.9958930217841712,
+      "sentence_nr": 0
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "hi",
@@ -35253,6 +40171,22 @@
       "score": 0.6376396416993303,
       "sentence_nr": 0
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "es",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.38318568210251663,
+      "sentence_nr": 0
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "es",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.6625830408830621,
+      "sentence_nr": 0
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "es",
@@ -35285,6 +40219,22 @@
       "score": 0.6399338911163,
       "sentence_nr": 0
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "es",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.42021658469726225,
+      "sentence_nr": 0
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "es",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.7002995337928327,
+      "sentence_nr": 0
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "es",
@@ -35429,6 +40379,22 @@
       "score": 0.6586847274336591,
       "sentence_nr": 0
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ar",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.668895061203786,
+      "sentence_nr": 0
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ar",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.7856645013611931,
+      "sentence_nr": 0
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ar",
@@ -35461,6 +40427,22 @@
       "score": 0.657598922173703,
       "sentence_nr": 0
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ar",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.731132155274915,
+      "sentence_nr": 0
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ar",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.8118306465406135,
+      "sentence_nr": 0
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ar",
@@ -35605,6 +40587,22 @@
       "score": 0.466691372759197,
       "sentence_nr": 0
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ur",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.3842600770501223,
+      "sentence_nr": 0
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ur",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.6030559477915464,
+      "sentence_nr": 0
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ur",
@@ -35637,6 +40635,22 @@
       "score": 0.46604753989124215,
       "sentence_nr": 0
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ur",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.36488083606907545,
+      "sentence_nr": 0
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ur",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.5811650865491297,
+      "sentence_nr": 0
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ur",
@@ -35781,6 +40795,22 @@
       "score": 0.6485553379227472,
       "sentence_nr": 0
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "fr",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.385626093679484,
+      "sentence_nr": 0
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "fr",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.6500036127408896,
+      "sentence_nr": 0
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "fr",
@@ -35813,6 +40843,22 @@
       "score": 0.6011484151165629,
       "sentence_nr": 0
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "fr",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.4326013853051836,
+      "sentence_nr": 0
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "fr",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.6609703936604562,
+      "sentence_nr": 0
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "fr",
@@ -35957,6 +41003,22 @@
       "score": 0.5972951640947346,
       "sentence_nr": 0
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "bn",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.37929353357736867,
+      "sentence_nr": 0
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "bn",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.6046831629978785,
+      "sentence_nr": 0
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "bn",
@@ -35989,6 +41051,22 @@
       "score": 0.6035836275599532,
       "sentence_nr": 0
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "bn",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.4839792901878845,
+      "sentence_nr": 0
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "bn",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.6380327835059832,
+      "sentence_nr": 0
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "bn",
@@ -36133,6 +41211,22 @@
       "score": 0.7060850657954441,
       "sentence_nr": 0
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pt",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.6832136298239752,
+      "sentence_nr": 0
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pt",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.7617777911358293,
+      "sentence_nr": 0
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "pt",
@@ -36165,6 +41259,22 @@
       "score": 0.7768522458527362,
       "sentence_nr": 0
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pt",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.6699094720554168,
+      "sentence_nr": 0
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pt",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.8022876242275274,
+      "sentence_nr": 0
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "pt",
@@ -36309,6 +41419,22 @@
       "score": 0.5906105668854662,
       "sentence_nr": 0
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pa",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.47662407876184354,
+      "sentence_nr": 0
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pa",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.5978297795375753,
+      "sentence_nr": 0
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "pa",
@@ -36341,6 +41467,22 @@
       "score": 0.5696298539086213,
       "sentence_nr": 0
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pa",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.44981308897808336,
+      "sentence_nr": 0
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pa",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.6151723374264357,
+      "sentence_nr": 0
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "pa",
@@ -36485,6 +41627,22 @@
       "score": 0.6348737822745005,
       "sentence_nr": 0
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ru",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.35291113737574475,
+      "sentence_nr": 0
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ru",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.6466347966614083,
+      "sentence_nr": 0
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ru",
@@ -36517,6 +41675,22 @@
       "score": 0.6897801155729197,
       "sentence_nr": 0
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ru",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.6036019309695121,
+      "sentence_nr": 0
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ru",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.7801644741948762,
+      "sentence_nr": 0
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ru",
@@ -36661,6 +41835,22 @@
       "score": 0.5073561650857479,
       "sentence_nr": 0
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "sw",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.40263336117444953,
+      "sentence_nr": 0
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "sw",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.656129562811693,
+      "sentence_nr": 0
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "sw",
@@ -36693,6 +41883,22 @@
       "score": 0.5336103118914343,
       "sentence_nr": 0
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "sw",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.4715421308516199,
+      "sentence_nr": 0
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "sw",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.7046296108422225,
+      "sentence_nr": 0
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "sw",
@@ -36837,6 +42043,22 @@
       "score": 0.7584810978753719,
       "sentence_nr": 0
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "id",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.3725685850787146,
+      "sentence_nr": 0
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "id",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.6944058070280722,
+      "sentence_nr": 0
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "id",
@@ -36869,6 +42091,22 @@
       "score": 0.6887265942100023,
       "sentence_nr": 0
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "id",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.4058364743511898,
+      "sentence_nr": 0
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "id",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.7590590041728482,
+      "sentence_nr": 0
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "id",
@@ -37013,6 +42251,22 @@
       "score": 0.6378039240510625,
       "sentence_nr": 0
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "de",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.400011517795393,
+      "sentence_nr": 0
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "de",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.6786441872043172,
+      "sentence_nr": 0
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "de",
@@ -37045,6 +42299,22 @@
       "score": 0.6426566938482169,
       "sentence_nr": 0
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "de",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.511936592363045,
+      "sentence_nr": 0
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "de",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.7671481459096147,
+      "sentence_nr": 0
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "de",
@@ -37189,6 +42459,22 @@
       "score": 0.003762227238525207,
       "sentence_nr": 0
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ja",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.30019255581073173,
+      "sentence_nr": 0
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ja",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.49869889490439867,
+      "sentence_nr": 0
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ja",
@@ -37221,6 +42507,22 @@
       "score": 0.4868090313482157,
       "sentence_nr": 0
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ja",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.4592260218476326,
+      "sentence_nr": 0
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ja",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.6081527520886763,
+      "sentence_nr": 0
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ja",
@@ -37365,6 +42667,22 @@
       "score": 0.7115011221714777,
       "sentence_nr": 1
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "en",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.305730257543728,
+      "sentence_nr": 1
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "en",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.6305034408922697,
+      "sentence_nr": 1
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "en",
@@ -37397,6 +42715,22 @@
       "score": 0.7089203664957927,
       "sentence_nr": 1
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "en",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.48186321118136805,
+      "sentence_nr": 1
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "en",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.7050536782800418,
+      "sentence_nr": 1
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "en",
@@ -37541,6 +42875,22 @@
       "score": 0.31119603942667584,
       "sentence_nr": 1
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "zh",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.3652139786200916,
+      "sentence_nr": 1
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "zh",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.3824205406345779,
+      "sentence_nr": 1
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "zh",
@@ -37573,6 +42923,22 @@
       "score": 0.3864890531682498,
       "sentence_nr": 1
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "zh",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.5702271610495845,
+      "sentence_nr": 1
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "zh",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.5772088119985683,
+      "sentence_nr": 1
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "zh",
@@ -37717,6 +43083,22 @@
       "score": 0.4497512968651573,
       "sentence_nr": 1
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "hi",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.1866741141650009,
+      "sentence_nr": 1
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "hi",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.46590330015791137,
+      "sentence_nr": 1
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "hi",
@@ -37749,6 +43131,22 @@
       "score": 0.4885052730214997,
       "sentence_nr": 1
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "hi",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.30632520148437686,
+      "sentence_nr": 1
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "hi",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.5904333377596244,
+      "sentence_nr": 1
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "hi",
@@ -37893,6 +43291,22 @@
       "score": 0.6358736384460296,
       "sentence_nr": 1
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "es",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.4310064928034671,
+      "sentence_nr": 1
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "es",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.6681893438144855,
+      "sentence_nr": 1
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "es",
@@ -37925,6 +43339,22 @@
       "score": 0.6319223068216205,
       "sentence_nr": 1
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "es",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.47486558569605275,
+      "sentence_nr": 1
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "es",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.6916379662719394,
+      "sentence_nr": 1
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "es",
@@ -38069,6 +43499,22 @@
       "score": 0.6662053431593723,
       "sentence_nr": 1
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ar",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.5216581079910853,
+      "sentence_nr": 1
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ar",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.71892039370478,
+      "sentence_nr": 1
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ar",
@@ -38101,6 +43547,22 @@
       "score": 0.6805608953669952,
       "sentence_nr": 1
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ar",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.5531306492249056,
+      "sentence_nr": 1
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ar",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.7508908077947477,
+      "sentence_nr": 1
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ar",
@@ -38245,6 +43707,22 @@
       "score": 0.3986479587107995,
       "sentence_nr": 1
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ur",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.2729012183957552,
+      "sentence_nr": 1
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ur",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.4275664216118961,
+      "sentence_nr": 1
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ur",
@@ -38277,6 +43755,22 @@
       "score": 0.3292454551002283,
       "sentence_nr": 1
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ur",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.0,
+      "sentence_nr": 1
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ur",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.09073980329024364,
+      "sentence_nr": 1
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ur",
@@ -38421,6 +43915,22 @@
       "score": 0.6853937472090788,
       "sentence_nr": 1
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "fr",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.5201554059039623,
+      "sentence_nr": 1
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "fr",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.6912283534522488,
+      "sentence_nr": 1
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "fr",
@@ -38453,6 +43963,22 @@
       "score": 0.6365023289177463,
       "sentence_nr": 1
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "fr",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.48149733895714797,
+      "sentence_nr": 1
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "fr",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.6355577992154319,
+      "sentence_nr": 1
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "fr",
@@ -38597,6 +44123,22 @@
       "score": 0.4853505495636382,
       "sentence_nr": 1
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "bn",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.25483341226864464,
+      "sentence_nr": 1
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "bn",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.4565768361133673,
+      "sentence_nr": 1
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "bn",
@@ -38629,6 +44171,22 @@
       "score": 0.47540661243586124,
       "sentence_nr": 1
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "bn",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.3721154325198309,
+      "sentence_nr": 1
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "bn",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.6214976185877734,
+      "sentence_nr": 1
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "bn",
@@ -38773,6 +44331,22 @@
       "score": 0.6584629522606407,
       "sentence_nr": 1
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pt",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.4767709962267032,
+      "sentence_nr": 1
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pt",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.6259826462063701,
+      "sentence_nr": 1
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "pt",
@@ -38805,6 +44379,22 @@
       "score": 0.5890799945028116,
       "sentence_nr": 1
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pt",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.49952715015218047,
+      "sentence_nr": 1
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pt",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.6422528647342378,
+      "sentence_nr": 1
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "pt",
@@ -38949,6 +44539,22 @@
       "score": 0.48318512703629857,
       "sentence_nr": 1
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pa",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.16510240061590087,
+      "sentence_nr": 1
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pa",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.37927849234648453,
+      "sentence_nr": 1
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "pa",
@@ -38981,6 +44587,22 @@
       "score": 0.3331532512757645,
       "sentence_nr": 1
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pa",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.469516870711376,
+      "sentence_nr": 1
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pa",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.6115882471919187,
+      "sentence_nr": 1
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "pa",
@@ -39125,6 +44747,22 @@
       "score": 0.6555816107178817,
       "sentence_nr": 1
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ru",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.38823222788076894,
+      "sentence_nr": 1
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ru",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.5684042820738361,
+      "sentence_nr": 1
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ru",
@@ -39157,6 +44795,22 @@
       "score": 0.6443063241500353,
       "sentence_nr": 1
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ru",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.6009332869110189,
+      "sentence_nr": 1
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ru",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.7201156007131091,
+      "sentence_nr": 1
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ru",
@@ -39301,6 +44955,22 @@
       "score": 0.6396787125445289,
       "sentence_nr": 1
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "sw",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.2589941364530944,
+      "sentence_nr": 1
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "sw",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.5932208556546101,
+      "sentence_nr": 1
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "sw",
@@ -39333,6 +45003,22 @@
       "score": 0.48559902973042135,
       "sentence_nr": 1
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "sw",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.4537382610431785,
+      "sentence_nr": 1
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "sw",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.6984640606230582,
+      "sentence_nr": 1
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "sw",
@@ -39477,6 +45163,22 @@
       "score": 0.5972281991369082,
       "sentence_nr": 1
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "id",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.33159620794115396,
+      "sentence_nr": 1
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "id",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.6420436747364323,
+      "sentence_nr": 1
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "id",
@@ -39509,6 +45211,22 @@
       "score": 0.632048088218684,
       "sentence_nr": 1
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "id",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.3728937115586142,
+      "sentence_nr": 1
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "id",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.6577808113928253,
+      "sentence_nr": 1
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "id",
@@ -39653,6 +45371,22 @@
       "score": 0.6258219368613708,
       "sentence_nr": 1
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "de",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.4876819889238188,
+      "sentence_nr": 1
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "de",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.6502930111654278,
+      "sentence_nr": 1
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "de",
@@ -39685,6 +45419,22 @@
       "score": 0.7135410085152943,
       "sentence_nr": 1
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "de",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.60059205595428,
+      "sentence_nr": 1
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "de",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.7065044125404809,
+      "sentence_nr": 1
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "de",
@@ -39829,6 +45579,22 @@
       "score": 0.2760884586584135,
       "sentence_nr": 1
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ja",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.2500271340971054,
+      "sentence_nr": 1
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ja",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.3785374220544885,
+      "sentence_nr": 1
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ja",
@@ -39861,6 +45627,22 @@
       "score": 0.37275082068184157,
       "sentence_nr": 1
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ja",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.3995850754493543,
+      "sentence_nr": 1
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ja",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.5087563569873156,
+      "sentence_nr": 1
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ja",
@@ -40005,6 +45787,22 @@
       "score": 0.6677259864784132,
       "sentence_nr": 2
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "en",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.6021903435236307,
+      "sentence_nr": 2
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "en",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.7189230568243182,
+      "sentence_nr": 2
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "en",
@@ -40037,6 +45835,22 @@
       "score": 0.6371098202414471,
       "sentence_nr": 2
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "en",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.7191192065120268,
+      "sentence_nr": 2
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "en",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.7900451518124424,
+      "sentence_nr": 2
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "en",
@@ -40181,6 +45995,22 @@
       "score": 0.12501819027374758,
       "sentence_nr": 2
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "zh",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.0,
+      "sentence_nr": 2
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "zh",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.1887796230243076,
+      "sentence_nr": 2
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "zh",
@@ -40213,6 +46043,22 @@
       "score": 0.22231961416584312,
       "sentence_nr": 2
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "zh",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.3487145358887869,
+      "sentence_nr": 2
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "zh",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.48564425156445185,
+      "sentence_nr": 2
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "zh",
@@ -40357,6 +46203,22 @@
       "score": 0.1485432117087218,
       "sentence_nr": 2
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "hi",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.12106878595868109,
+      "sentence_nr": 2
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "hi",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.3192454012719998,
+      "sentence_nr": 2
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "hi",
@@ -40389,6 +46251,22 @@
       "score": 0.388678103641788,
       "sentence_nr": 2
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "hi",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.3259738911297118,
+      "sentence_nr": 2
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "hi",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.4954841706551886,
+      "sentence_nr": 2
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "hi",
@@ -40533,6 +46411,22 @@
       "score": 0.5012707040525209,
       "sentence_nr": 2
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "es",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.6297960258710876,
+      "sentence_nr": 2
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "es",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.7117676662366008,
+      "sentence_nr": 2
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "es",
@@ -40565,6 +46459,22 @@
       "score": 0.4827709277987172,
       "sentence_nr": 2
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "es",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.3947812939950854,
+      "sentence_nr": 2
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "es",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.5373384617862703,
+      "sentence_nr": 2
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "es",
@@ -40709,6 +46619,22 @@
       "score": 0.46792167630295967,
       "sentence_nr": 2
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ar",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.37475590201160436,
+      "sentence_nr": 2
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ar",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.549178868228435,
+      "sentence_nr": 2
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ar",
@@ -40741,6 +46667,22 @@
       "score": 0.2986174009048306,
       "sentence_nr": 2
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ar",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.4176763688729275,
+      "sentence_nr": 2
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ar",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.6124700716856484,
+      "sentence_nr": 2
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ar",
@@ -40885,6 +46827,22 @@
       "score": 0.24480102898506534,
       "sentence_nr": 2
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ur",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.0,
+      "sentence_nr": 2
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ur",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.2623806581920467,
+      "sentence_nr": 2
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ur",
@@ -40917,6 +46875,22 @@
       "score": 0.1582263258709324,
       "sentence_nr": 2
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ur",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.3184011333042053,
+      "sentence_nr": 2
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ur",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.475779589579386,
+      "sentence_nr": 2
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ur",
@@ -41061,6 +47035,22 @@
       "score": 0.575799986766,
       "sentence_nr": 2
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "fr",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.3877240689639599,
+      "sentence_nr": 2
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "fr",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.5570859361697285,
+      "sentence_nr": 2
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "fr",
@@ -41093,6 +47083,22 @@
       "score": 0.44291475401588093,
       "sentence_nr": 2
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "fr",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.40977628070530747,
+      "sentence_nr": 2
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "fr",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.5623921018498135,
+      "sentence_nr": 2
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "fr",
@@ -41237,6 +47243,22 @@
       "score": 0.23021641289829473,
       "sentence_nr": 2
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "bn",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.3297358210077752,
+      "sentence_nr": 2
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "bn",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.506224157487821,
+      "sentence_nr": 2
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "bn",
@@ -41269,6 +47291,22 @@
       "score": 0.2323275601638909,
       "sentence_nr": 2
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "bn",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.4592557039164775,
+      "sentence_nr": 2
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "bn",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.6121399518555048,
+      "sentence_nr": 2
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "bn",
@@ -41413,6 +47451,22 @@
       "score": 0.571328063702761,
       "sentence_nr": 2
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pt",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.6077585258730265,
+      "sentence_nr": 2
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pt",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.7400152006566423,
+      "sentence_nr": 2
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "pt",
@@ -41445,6 +47499,22 @@
       "score": 0.5022745285039809,
       "sentence_nr": 2
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pt",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.5803184114968359,
+      "sentence_nr": 2
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pt",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.759870828515734,
+      "sentence_nr": 2
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "pt",
@@ -41589,6 +47659,22 @@
       "score": 0.15363234192450648,
       "sentence_nr": 2
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pa",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.0,
+      "sentence_nr": 2
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pa",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.251174652769276,
+      "sentence_nr": 2
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "pa",
@@ -41621,6 +47707,22 @@
       "score": 0.13978782442553714,
       "sentence_nr": 2
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pa",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.1898569805320716,
+      "sentence_nr": 2
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pa",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.3803223837566382,
+      "sentence_nr": 2
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "pa",
@@ -41765,6 +47867,22 @@
       "score": 0.5643748237802169,
       "sentence_nr": 2
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ru",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.40332947519159895,
+      "sentence_nr": 2
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ru",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.6228862504867423,
+      "sentence_nr": 2
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ru",
@@ -41797,6 +47915,22 @@
       "score": 0.5618587771651018,
       "sentence_nr": 2
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ru",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.43155890347066467,
+      "sentence_nr": 2
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ru",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.619009447565164,
+      "sentence_nr": 2
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ru",
@@ -41941,6 +48075,22 @@
       "score": 0.5061357551531296,
       "sentence_nr": 2
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "sw",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.3958704329397872,
+      "sentence_nr": 2
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "sw",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.5196505496421775,
+      "sentence_nr": 2
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "sw",
@@ -41973,6 +48123,22 @@
       "score": 0.49086962788909555,
       "sentence_nr": 2
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "sw",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.5117916534946495,
+      "sentence_nr": 2
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "sw",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.6878657354957926,
+      "sentence_nr": 2
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "sw",
@@ -42117,6 +48283,22 @@
       "score": 0.5475770179024447,
       "sentence_nr": 2
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "id",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.5354397296450966,
+      "sentence_nr": 2
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "id",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.6627191681525589,
+      "sentence_nr": 2
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "id",
@@ -42149,6 +48331,22 @@
       "score": 0.5818873909634904,
       "sentence_nr": 2
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "id",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.674363352915248,
+      "sentence_nr": 2
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "id",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.7903753469468177,
+      "sentence_nr": 2
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "id",
@@ -42293,6 +48491,22 @@
       "score": 0.43207538722163397,
       "sentence_nr": 2
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "de",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.1148153812028893,
+      "sentence_nr": 2
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "de",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.20549791555765032,
+      "sentence_nr": 2
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "de",
@@ -42325,6 +48539,22 @@
       "score": 0.4061072764676573,
       "sentence_nr": 2
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "de",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.355480478065782,
+      "sentence_nr": 2
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "de",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.5288950976571054,
+      "sentence_nr": 2
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "de",
@@ -42469,6 +48699,22 @@
       "score": 0.13084917716157846,
       "sentence_nr": 2
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ja",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.44404782758976735,
+      "sentence_nr": 2
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ja",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.5596520713104719,
+      "sentence_nr": 2
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ja",
@@ -42501,6 +48747,22 @@
       "score": 0.3931394603219493,
       "sentence_nr": 2
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ja",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.2977932227461586,
+      "sentence_nr": 2
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ja",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.45438903925756086,
+      "sentence_nr": 2
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ja",
@@ -42645,6 +48907,22 @@
       "score": 0.7558874882119336,
       "sentence_nr": 3
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "en",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.24688498672025874,
+      "sentence_nr": 3
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "en",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.6300112897041039,
+      "sentence_nr": 3
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "en",
@@ -42677,6 +48955,22 @@
       "score": 0.9484564543183253,
       "sentence_nr": 3
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "en",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 1.0,
+      "sentence_nr": 3
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "en",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 1.0,
+      "sentence_nr": 3
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "en",
@@ -42821,6 +49115,22 @@
       "score": 0.10947303419437356,
       "sentence_nr": 3
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "zh",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.0,
+      "sentence_nr": 3
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "zh",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.09122941759796505,
+      "sentence_nr": 3
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "zh",
@@ -42853,6 +49163,22 @@
       "score": 0.11147384852362276,
       "sentence_nr": 3
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "zh",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.0,
+      "sentence_nr": 3
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "zh",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.07933317425857943,
+      "sentence_nr": 3
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "zh",
@@ -42997,6 +49323,22 @@
       "score": 0.6643984252563968,
       "sentence_nr": 3
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "hi",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.760856626273165,
+      "sentence_nr": 3
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "hi",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.881580297011256,
+      "sentence_nr": 3
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "hi",
@@ -43029,6 +49371,22 @@
       "score": 0.7539119883011114,
       "sentence_nr": 3
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "hi",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.5824621545691198,
+      "sentence_nr": 3
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "hi",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.6375144448777752,
+      "sentence_nr": 3
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "hi",
@@ -43173,6 +49531,22 @@
       "score": 0.6732018003142922,
       "sentence_nr": 3
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "es",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.26529518334824453,
+      "sentence_nr": 3
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "es",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.596699960316198,
+      "sentence_nr": 3
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "es",
@@ -43205,6 +49579,22 @@
       "score": 0.6117499551501043,
       "sentence_nr": 3
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "es",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.4747354911173249,
+      "sentence_nr": 3
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "es",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.7360286800047513,
+      "sentence_nr": 3
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "es",
@@ -43349,6 +49739,22 @@
       "score": 0.26860011657329247,
       "sentence_nr": 3
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ar",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.1435622311718879,
+      "sentence_nr": 3
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ar",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.45714671497681403,
+      "sentence_nr": 3
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ar",
@@ -43381,6 +49787,22 @@
       "score": 0.4800955244005148,
       "sentence_nr": 3
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ar",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.18482936243672016,
+      "sentence_nr": 3
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ar",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.5172215726655364,
+      "sentence_nr": 3
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ar",
@@ -43525,6 +49947,22 @@
       "score": 0.320678468026793,
       "sentence_nr": 3
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ur",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.3124325727595954,
+      "sentence_nr": 3
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ur",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.4415757258745415,
+      "sentence_nr": 3
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ur",
@@ -43557,6 +49995,22 @@
       "score": 0.3630016390465325,
       "sentence_nr": 3
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ur",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.21850594525107195,
+      "sentence_nr": 3
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ur",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.4173980390626746,
+      "sentence_nr": 3
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ur",
@@ -43701,6 +50155,22 @@
       "score": 0.4761746966391582,
       "sentence_nr": 3
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "fr",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.1935951733925871,
+      "sentence_nr": 3
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "fr",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.43667702869251973,
+      "sentence_nr": 3
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "fr",
@@ -43733,6 +50203,22 @@
       "score": 0.684329671666446,
       "sentence_nr": 3
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "fr",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.8787419089273848,
+      "sentence_nr": 3
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "fr",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.9422733087334002,
+      "sentence_nr": 3
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "fr",
@@ -43877,6 +50363,22 @@
       "score": 0.433708341935832,
       "sentence_nr": 3
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "bn",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.3734832062562986,
+      "sentence_nr": 3
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "bn",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.44543323722408085,
+      "sentence_nr": 3
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "bn",
@@ -43909,6 +50411,22 @@
       "score": 0.43422338821405304,
       "sentence_nr": 3
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "bn",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.5894159589207006,
+      "sentence_nr": 3
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "bn",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.633437763335489,
+      "sentence_nr": 3
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "bn",
@@ -44053,6 +50571,22 @@
       "score": 0.6528501353073614,
       "sentence_nr": 3
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pt",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.26697411956933875,
+      "sentence_nr": 3
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pt",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.6328956554106696,
+      "sentence_nr": 3
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "pt",
@@ -44085,6 +50619,22 @@
       "score": 0.6888233111124319,
       "sentence_nr": 3
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pt",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.670001214025099,
+      "sentence_nr": 3
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pt",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.876386276114813,
+      "sentence_nr": 3
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "pt",
@@ -44229,6 +50779,22 @@
       "score": 0.0,
       "sentence_nr": 3
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pa",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 1.0,
+      "sentence_nr": 3
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pa",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 1.0,
+      "sentence_nr": 3
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "pa",
@@ -44261,6 +50827,22 @@
       "score": 1.0,
       "sentence_nr": 3
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pa",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 1.0,
+      "sentence_nr": 3
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pa",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 1.0,
+      "sentence_nr": 3
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "pa",
@@ -44405,6 +50987,22 @@
       "score": 0.3821582738832969,
       "sentence_nr": 3
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ru",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.21171273476282318,
+      "sentence_nr": 3
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ru",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.34680801952866847,
+      "sentence_nr": 3
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ru",
@@ -44437,6 +51035,22 @@
       "score": 0.4039853784752083,
       "sentence_nr": 3
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ru",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.4081354056739722,
+      "sentence_nr": 3
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ru",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.47955174657690236,
+      "sentence_nr": 3
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ru",
@@ -44581,6 +51195,22 @@
       "score": 0.5427832684043266,
       "sentence_nr": 3
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "sw",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.0993195473228234,
+      "sentence_nr": 3
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "sw",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.4660581946805371,
+      "sentence_nr": 3
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "sw",
@@ -44613,6 +51243,22 @@
       "score": 0.37306669253790053,
       "sentence_nr": 3
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "sw",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.866397551781362,
+      "sentence_nr": 3
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "sw",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.9369949537059603,
+      "sentence_nr": 3
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "sw",
@@ -44757,6 +51403,22 @@
       "score": 0.6399673599980337,
       "sentence_nr": 3
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "id",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.37251337991409605,
+      "sentence_nr": 3
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "id",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.5810872572798261,
+      "sentence_nr": 3
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "id",
@@ -44789,6 +51451,22 @@
       "score": 0.648844691127488,
       "sentence_nr": 3
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "id",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.33084780351073634,
+      "sentence_nr": 3
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "id",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.5879159712556987,
+      "sentence_nr": 3
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "id",
@@ -44933,6 +51611,22 @@
       "score": 0.4267826722481737,
       "sentence_nr": 3
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "de",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.18331704949485053,
+      "sentence_nr": 3
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "de",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.4681674930025697,
+      "sentence_nr": 3
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "de",
@@ -44965,6 +51659,22 @@
       "score": 0.5956068369645927,
       "sentence_nr": 3
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "de",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.4546852631699836,
+      "sentence_nr": 3
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "de",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.7654091839521726,
+      "sentence_nr": 3
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "de",
@@ -45109,6 +51819,22 @@
       "score": 0.27080894796384963,
       "sentence_nr": 3
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ja",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.21800193956058223,
+      "sentence_nr": 3
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ja",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.2974641182469979,
+      "sentence_nr": 3
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ja",
@@ -45141,6 +51867,22 @@
       "score": 0.3517185856118227,
       "sentence_nr": 3
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ja",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.25306188056493334,
+      "sentence_nr": 3
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ja",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.3181932375842872,
+      "sentence_nr": 3
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ja",
@@ -45285,6 +52027,22 @@
       "score": 0.2246029757863831,
       "sentence_nr": 4
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "en",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 1.0,
+      "sentence_nr": 4
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "en",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 1.0,
+      "sentence_nr": 4
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "en",
@@ -45317,6 +52075,22 @@
       "score": 0.9453478043428296,
       "sentence_nr": 4
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "en",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 1.0,
+      "sentence_nr": 4
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "en",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 1.0,
+      "sentence_nr": 4
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "en",
@@ -45461,6 +52235,22 @@
       "score": 0.3482814151315599,
       "sentence_nr": 4
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "zh",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.6219720158712322,
+      "sentence_nr": 4
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "zh",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.638758039725182,
+      "sentence_nr": 4
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "zh",
@@ -45493,6 +52283,22 @@
       "score": 0.7205373993220106,
       "sentence_nr": 4
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "zh",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.611843760819802,
+      "sentence_nr": 4
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "zh",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.6194911484836914,
+      "sentence_nr": 4
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "zh",
@@ -45637,6 +52443,22 @@
       "score": 0.2850647115160651,
       "sentence_nr": 4
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "hi",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.0,
+      "sentence_nr": 4
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "hi",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.3175465093373464,
+      "sentence_nr": 4
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "hi",
@@ -45669,6 +52491,22 @@
       "score": 0.4033902612785559,
       "sentence_nr": 4
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "hi",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.13598345546333285,
+      "sentence_nr": 4
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "hi",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.4135396704381328,
+      "sentence_nr": 4
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "hi",
@@ -45813,6 +52651,22 @@
       "score": 0.5392658386159207,
       "sentence_nr": 4
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "es",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.2293530951556094,
+      "sentence_nr": 4
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "es",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.4689601314620498,
+      "sentence_nr": 4
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "es",
@@ -45845,6 +52699,22 @@
       "score": 0.5488313413379253,
       "sentence_nr": 4
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "es",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.3820562306791339,
+      "sentence_nr": 4
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "es",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.5508290063627067,
+      "sentence_nr": 4
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "es",
@@ -45989,6 +52859,22 @@
       "score": 0.4093450185788297,
       "sentence_nr": 4
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ar",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.23274285105688466,
+      "sentence_nr": 4
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ar",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.4091360043642453,
+      "sentence_nr": 4
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ar",
@@ -46021,6 +52907,22 @@
       "score": 0.37850093315889116,
       "sentence_nr": 4
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ar",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.23251355381714656,
+      "sentence_nr": 4
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ar",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.42986991593991275,
+      "sentence_nr": 4
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ar",
@@ -46165,6 +53067,22 @@
       "score": 0.3743678965131091,
       "sentence_nr": 4
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ur",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.15511550090520096,
+      "sentence_nr": 4
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ur",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.3922184662482167,
+      "sentence_nr": 4
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ur",
@@ -46197,6 +53115,22 @@
       "score": 0.2957279302594959,
       "sentence_nr": 4
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ur",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.10793583834163357,
+      "sentence_nr": 4
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ur",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.3079497311888636,
+      "sentence_nr": 4
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ur",
@@ -46341,6 +53275,22 @@
       "score": 0.5108598154804425,
       "sentence_nr": 4
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "fr",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.32163989714697483,
+      "sentence_nr": 4
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "fr",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.4414323713243047,
+      "sentence_nr": 4
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "fr",
@@ -46373,6 +53323,22 @@
       "score": 0.5361569875660316,
       "sentence_nr": 4
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "fr",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.20913838136220486,
+      "sentence_nr": 4
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "fr",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.40960094031121963,
+      "sentence_nr": 4
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "fr",
@@ -46517,6 +53483,22 @@
       "score": 0.48172049854477195,
       "sentence_nr": 4
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "bn",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.22872196013470597,
+      "sentence_nr": 4
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "bn",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.45089857576633846,
+      "sentence_nr": 4
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "bn",
@@ -46549,6 +53531,22 @@
       "score": 0.34629467658248214,
       "sentence_nr": 4
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "bn",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.17550354183836317,
+      "sentence_nr": 4
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "bn",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.3977547521857469,
+      "sentence_nr": 4
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "bn",
@@ -46693,6 +53691,22 @@
       "score": 0.6087635830564418,
       "sentence_nr": 4
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pt",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.13836903384315105,
+      "sentence_nr": 4
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pt",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.5516548411073219,
+      "sentence_nr": 4
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "pt",
@@ -46725,6 +53739,22 @@
       "score": 0.5109257435313587,
       "sentence_nr": 4
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pt",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.34589895849033103,
+      "sentence_nr": 4
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pt",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.6136307264772042,
+      "sentence_nr": 4
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "pt",
@@ -46869,6 +53899,22 @@
       "score": 0.3385577201847465,
       "sentence_nr": 4
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pa",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.11556653761629153,
+      "sentence_nr": 4
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pa",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.3140069931838876,
+      "sentence_nr": 4
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "pa",
@@ -46901,6 +53947,22 @@
       "score": 0.2933944065312711,
       "sentence_nr": 4
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pa",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.1055629358593665,
+      "sentence_nr": 4
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pa",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.30482671461107386,
+      "sentence_nr": 4
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "pa",
@@ -47045,6 +54107,22 @@
       "score": 0.6999124430022288,
       "sentence_nr": 4
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ru",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.5276151436342643,
+      "sentence_nr": 4
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ru",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.6871514991080862,
+      "sentence_nr": 4
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ru",
@@ -47077,6 +54155,22 @@
       "score": 0.7471196627888963,
       "sentence_nr": 4
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ru",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.5274964121279998,
+      "sentence_nr": 4
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ru",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.6941952618694388,
+      "sentence_nr": 4
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ru",
@@ -47221,6 +54315,22 @@
       "score": 0.43519517439687405,
       "sentence_nr": 4
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "sw",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.32705341718250747,
+      "sentence_nr": 4
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "sw",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.631354608587055,
+      "sentence_nr": 4
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "sw",
@@ -47253,6 +54363,22 @@
       "score": 0.5288881528593262,
       "sentence_nr": 4
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "sw",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.5831654195482086,
+      "sentence_nr": 4
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "sw",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.7601444481236787,
+      "sentence_nr": 4
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "sw",
@@ -47397,6 +54523,22 @@
       "score": 0.4692685009782657,
       "sentence_nr": 4
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "id",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.2680165156355779,
+      "sentence_nr": 4
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "id",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.5590529072823445,
+      "sentence_nr": 4
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "id",
@@ -47429,6 +54571,22 @@
       "score": 0.5563318425026342,
       "sentence_nr": 4
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "id",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.3200938205435179,
+      "sentence_nr": 4
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "id",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.6393945963053513,
+      "sentence_nr": 4
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "id",
@@ -47573,6 +54731,22 @@
       "score": 0.5926217012511299,
       "sentence_nr": 4
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "de",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.30470915491420003,
+      "sentence_nr": 4
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "de",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.542875812219914,
+      "sentence_nr": 4
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "de",
@@ -47605,6 +54779,22 @@
       "score": 0.611702219968759,
       "sentence_nr": 4
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "de",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.312793730905921,
+      "sentence_nr": 4
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "de",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.4874766693766197,
+      "sentence_nr": 4
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "de",
@@ -47749,6 +54939,22 @@
       "score": 0.0,
       "sentence_nr": 4
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ja",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.1845895819969781,
+      "sentence_nr": 4
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ja",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.33578716975410133,
+      "sentence_nr": 4
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ja",
@@ -47781,6 +54987,22 @@
       "score": 0.28916309026824916,
       "sentence_nr": 4
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ja",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.6681082569496674,
+      "sentence_nr": 4
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ja",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.7257478469803625,
+      "sentence_nr": 4
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ja",
@@ -47925,6 +55147,22 @@
       "score": 0.2381658499765768,
       "sentence_nr": 5
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "en",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 1.0,
+      "sentence_nr": 5
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "en",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 1.0,
+      "sentence_nr": 5
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "en",
@@ -47957,6 +55195,22 @@
       "score": 0.9063898435384111,
       "sentence_nr": 5
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "en",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 1.0,
+      "sentence_nr": 5
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "en",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 1.0,
+      "sentence_nr": 5
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "en",
@@ -48101,6 +55355,22 @@
       "score": 0.314589204347422,
       "sentence_nr": 5
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "zh",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.31383720140423793,
+      "sentence_nr": 5
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "zh",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.35896609082765174,
+      "sentence_nr": 5
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "zh",
@@ -48133,6 +55403,22 @@
       "score": 0.3462132320098601,
       "sentence_nr": 5
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "zh",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.30613574556266654,
+      "sentence_nr": 5
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "zh",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.36162356523761796,
+      "sentence_nr": 5
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "zh",
@@ -48277,6 +55563,22 @@
       "score": 0.7669087484597642,
       "sentence_nr": 5
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "hi",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.6153267326643309,
+      "sentence_nr": 5
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "hi",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.825440708536967,
+      "sentence_nr": 5
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "hi",
@@ -48309,6 +55611,22 @@
       "score": 0.7523344918083558,
       "sentence_nr": 5
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "hi",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.46997395980026974,
+      "sentence_nr": 5
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "hi",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.8114935753258365,
+      "sentence_nr": 5
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "hi",
@@ -48453,6 +55771,22 @@
       "score": 0.5552412314880962,
       "sentence_nr": 5
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "es",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.6976333495952621,
+      "sentence_nr": 5
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "es",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.8331572107884448,
+      "sentence_nr": 5
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "es",
@@ -48485,6 +55819,22 @@
       "score": 0.6664090181705107,
       "sentence_nr": 5
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "es",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.5440766840557734,
+      "sentence_nr": 5
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "es",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.7530101164980872,
+      "sentence_nr": 5
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "es",
@@ -48629,6 +55979,22 @@
       "score": 0.48645628248697975,
       "sentence_nr": 5
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ar",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.498704623570665,
+      "sentence_nr": 5
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ar",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.6478746389895599,
+      "sentence_nr": 5
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ar",
@@ -48661,6 +56027,22 @@
       "score": 0.4447177675003817,
       "sentence_nr": 5
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ar",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.498704623570665,
+      "sentence_nr": 5
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ar",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.6478746389895599,
+      "sentence_nr": 5
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ar",
@@ -48805,6 +56187,22 @@
       "score": 0.4729753929525169,
       "sentence_nr": 5
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ur",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.2622716439052442,
+      "sentence_nr": 5
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ur",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.42384915893461766,
+      "sentence_nr": 5
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ur",
@@ -48837,6 +56235,22 @@
       "score": 0.38732841080078323,
       "sentence_nr": 5
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ur",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.30690336937373786,
+      "sentence_nr": 5
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ur",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.46380735910652676,
+      "sentence_nr": 5
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ur",
@@ -48981,6 +56395,22 @@
       "score": 0.7639130574395125,
       "sentence_nr": 5
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "fr",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.833078701050083,
+      "sentence_nr": 5
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "fr",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.9482515348146272,
+      "sentence_nr": 5
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "fr",
@@ -49013,6 +56443,22 @@
       "score": 0.8158797976578578,
       "sentence_nr": 5
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "fr",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 1.0,
+      "sentence_nr": 5
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "fr",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 1.0,
+      "sentence_nr": 5
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "fr",
@@ -49157,6 +56603,22 @@
       "score": 0.4017565065239436,
       "sentence_nr": 5
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "bn",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.4933292241270431,
+      "sentence_nr": 5
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "bn",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.511824430191619,
+      "sentence_nr": 5
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "bn",
@@ -49189,6 +56651,22 @@
       "score": 0.41380245501613677,
       "sentence_nr": 5
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "bn",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.49539605131242165,
+      "sentence_nr": 5
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "bn",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.44728880966754114,
+      "sentence_nr": 5
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "bn",
@@ -49333,6 +56811,22 @@
       "score": 0.5095895501997145,
       "sentence_nr": 5
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pt",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.49961369350177665,
+      "sentence_nr": 5
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pt",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.6351045115684573,
+      "sentence_nr": 5
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "pt",
@@ -49365,6 +56859,22 @@
       "score": 0.5822934956325967,
       "sentence_nr": 5
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pt",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.49885432872486163,
+      "sentence_nr": 5
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pt",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.6436097366017006,
+      "sentence_nr": 5
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "pt",
@@ -49509,6 +57019,22 @@
       "score": 0.5800761309604682,
       "sentence_nr": 5
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pa",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.35138749399652214,
+      "sentence_nr": 5
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pa",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.5704306601285767,
+      "sentence_nr": 5
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "pa",
@@ -49541,6 +57067,22 @@
       "score": 0.5691358329649412,
       "sentence_nr": 5
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pa",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.46550355389682435,
+      "sentence_nr": 5
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pa",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.646730632916617,
+      "sentence_nr": 5
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "pa",
@@ -49685,6 +57227,22 @@
       "score": 0.518761522736185,
       "sentence_nr": 5
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ru",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.0,
+      "sentence_nr": 5
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ru",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.518761522736185,
+      "sentence_nr": 5
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ru",
@@ -49717,6 +57275,22 @@
       "score": 0.4397936463531347,
       "sentence_nr": 5
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ru",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.39814417587130846,
+      "sentence_nr": 5
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ru",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.6510906999464993,
+      "sentence_nr": 5
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ru",
@@ -49861,6 +57435,22 @@
       "score": 0.5564992960428438,
       "sentence_nr": 5
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "sw",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.44778459441351737,
+      "sentence_nr": 5
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "sw",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.7043336945393497,
+      "sentence_nr": 5
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "sw",
@@ -49893,6 +57483,22 @@
       "score": 0.7511716303980656,
       "sentence_nr": 5
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "sw",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.5106109398471469,
+      "sentence_nr": 5
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "sw",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.7688046995197549,
+      "sentence_nr": 5
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "sw",
@@ -50037,6 +57643,22 @@
       "score": 0.8116199676115453,
       "sentence_nr": 5
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "id",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.6535194995338728,
+      "sentence_nr": 5
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "id",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.8909391457425937,
+      "sentence_nr": 5
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "id",
@@ -50069,6 +57691,22 @@
       "score": 0.8116199676115453,
       "sentence_nr": 5
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "id",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.5088645484558708,
+      "sentence_nr": 5
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "id",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.8116199676115453,
+      "sentence_nr": 5
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "id",
@@ -50213,6 +57851,22 @@
       "score": 0.6707532211471023,
       "sentence_nr": 5
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "de",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.7298378378464025,
+      "sentence_nr": 5
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "de",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.8542310686470678,
+      "sentence_nr": 5
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "de",
@@ -50245,6 +57899,22 @@
       "score": 0.7147018027438421,
       "sentence_nr": 5
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "de",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.7483293841345244,
+      "sentence_nr": 5
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "de",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.8146526693270999,
+      "sentence_nr": 5
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "de",
@@ -50389,6 +58059,22 @@
       "score": 0.01834337391695103,
       "sentence_nr": 5
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ja",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.1540231640374204,
+      "sentence_nr": 5
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ja",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.36182698586213136,
+      "sentence_nr": 5
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ja",
@@ -50421,6 +58107,22 @@
       "score": 0.4432072463778114,
       "sentence_nr": 5
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ja",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.3682241310101735,
+      "sentence_nr": 5
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ja",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.5450474312451057,
+      "sentence_nr": 5
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ja",
@@ -50565,6 +58267,22 @@
       "score": 0.3263040636562357,
       "sentence_nr": 6
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "en",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.7511573912724299,
+      "sentence_nr": 6
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "en",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.9453473543978153,
+      "sentence_nr": 6
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "en",
@@ -50597,6 +58315,22 @@
       "score": 0.9453473543978153,
       "sentence_nr": 6
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "en",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.7511573912724299,
+      "sentence_nr": 6
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "en",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.9453473543978153,
+      "sentence_nr": 6
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "en",
@@ -50741,6 +58475,22 @@
       "score": 0.5243375045345786,
       "sentence_nr": 6
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "zh",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.328872375046221,
+      "sentence_nr": 6
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "zh",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.39142733167976973,
+      "sentence_nr": 6
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "zh",
@@ -50773,6 +58523,22 @@
       "score": 0.28295274449167956,
       "sentence_nr": 6
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "zh",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.3567464687150701,
+      "sentence_nr": 6
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "zh",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.3651755892066728,
+      "sentence_nr": 6
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "zh",
@@ -50917,6 +58683,22 @@
       "score": 0.5076725973953424,
       "sentence_nr": 6
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "hi",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.29300728994404895,
+      "sentence_nr": 6
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "hi",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.4620516485185446,
+      "sentence_nr": 6
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "hi",
@@ -50949,6 +58731,22 @@
       "score": 0.5497265770945076,
       "sentence_nr": 6
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "hi",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.4544505188404086,
+      "sentence_nr": 6
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "hi",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.6033006987520517,
+      "sentence_nr": 6
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "hi",
@@ -51093,6 +58891,22 @@
       "score": 0.7317828775912516,
       "sentence_nr": 6
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "es",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.5521710658453207,
+      "sentence_nr": 6
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "es",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.7317828775912516,
+      "sentence_nr": 6
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "es",
@@ -51125,6 +58939,22 @@
       "score": 0.7317828775912516,
       "sentence_nr": 6
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "es",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.5521710658453207,
+      "sentence_nr": 6
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "es",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.7317828775912516,
+      "sentence_nr": 6
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "es",
@@ -51269,6 +59099,22 @@
       "score": 0.526357446896968,
       "sentence_nr": 6
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ar",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.5582486914071635,
+      "sentence_nr": 6
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ar",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.6678493404097802,
+      "sentence_nr": 6
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ar",
@@ -51301,6 +59147,22 @@
       "score": 0.7571125338649978,
       "sentence_nr": 6
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ar",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.5406438522344627,
+      "sentence_nr": 6
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ar",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.6520694800788391,
+      "sentence_nr": 6
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ar",
@@ -51445,6 +59307,22 @@
       "score": 0.466645869611307,
       "sentence_nr": 6
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ur",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.20422838465921236,
+      "sentence_nr": 6
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ur",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.3821382671218279,
+      "sentence_nr": 6
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ur",
@@ -51477,6 +59355,22 @@
       "score": 0.45184360988354105,
       "sentence_nr": 6
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ur",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.2599542517888651,
+      "sentence_nr": 6
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ur",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.49098929416640624,
+      "sentence_nr": 6
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ur",
@@ -51621,6 +59515,22 @@
       "score": 0.9495292423959529,
       "sentence_nr": 6
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "fr",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 1.0,
+      "sentence_nr": 6
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "fr",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 1.0,
+      "sentence_nr": 6
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "fr",
@@ -51653,6 +59563,22 @@
       "score": 0.8078891929749037,
       "sentence_nr": 6
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "fr",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.7629273292796576,
+      "sentence_nr": 6
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "fr",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.8510385544954956,
+      "sentence_nr": 6
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "fr",
@@ -51797,6 +59723,22 @@
       "score": 0.5217663812589132,
       "sentence_nr": 6
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "bn",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.18189587992135597,
+      "sentence_nr": 6
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "bn",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.43964080400724653,
+      "sentence_nr": 6
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "bn",
@@ -51829,6 +59771,22 @@
       "score": 0.5506555496793699,
       "sentence_nr": 6
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "bn",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.6495308560002527,
+      "sentence_nr": 6
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "bn",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.7114963534919011,
+      "sentence_nr": 6
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "bn",
@@ -51973,6 +59931,22 @@
       "score": 0.7679844670813416,
       "sentence_nr": 6
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pt",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.8253498772794055,
+      "sentence_nr": 6
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pt",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.8529564805429163,
+      "sentence_nr": 6
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "pt",
@@ -52005,6 +59979,22 @@
       "score": 1.0,
       "sentence_nr": 6
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pt",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.7072172847953276,
+      "sentence_nr": 6
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pt",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.7914639887327892,
+      "sentence_nr": 6
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "pt",
@@ -52149,6 +60139,22 @@
       "score": 0.4999323991212311,
       "sentence_nr": 6
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pa",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.26332019392396333,
+      "sentence_nr": 6
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pa",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.5893220054460814,
+      "sentence_nr": 6
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "pa",
@@ -52181,6 +60187,22 @@
       "score": 0.5730023382770898,
       "sentence_nr": 6
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pa",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.4161791450287817,
+      "sentence_nr": 6
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pa",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.6046852394200818,
+      "sentence_nr": 6
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "pa",
@@ -52325,6 +60347,22 @@
       "score": 0.5501364764829885,
       "sentence_nr": 6
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ru",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.4369942407063455,
+      "sentence_nr": 6
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ru",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.6009573115008555,
+      "sentence_nr": 6
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ru",
@@ -52357,6 +60395,22 @@
       "score": 0.5631437828635808,
       "sentence_nr": 6
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ru",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.7498810286408993,
+      "sentence_nr": 6
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ru",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.7726337964681356,
+      "sentence_nr": 6
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ru",
@@ -52501,6 +60555,22 @@
       "score": 0.5721164465661742,
       "sentence_nr": 6
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "sw",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.15495096883986592,
+      "sentence_nr": 6
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "sw",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.45421263258392414,
+      "sentence_nr": 6
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "sw",
@@ -52533,6 +60603,22 @@
       "score": 0.42099734580654347,
       "sentence_nr": 6
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "sw",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.24518299917312847,
+      "sentence_nr": 6
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "sw",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.5623282316325473,
+      "sentence_nr": 6
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "sw",
@@ -52677,6 +60763,22 @@
       "score": 0.5159811845433955,
       "sentence_nr": 6
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "id",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.35504200505176187,
+      "sentence_nr": 6
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "id",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.5693079918450474,
+      "sentence_nr": 6
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "id",
@@ -52709,6 +60811,22 @@
       "score": 0.45963072970927465,
       "sentence_nr": 6
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "id",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.1794560313432444,
+      "sentence_nr": 6
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "id",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.5236301264596329,
+      "sentence_nr": 6
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "id",
@@ -52853,6 +60971,22 @@
       "score": 0.7887116805325072,
       "sentence_nr": 6
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "de",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.6325987025085013,
+      "sentence_nr": 6
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "de",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.7966336957924106,
+      "sentence_nr": 6
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "de",
@@ -52885,6 +61019,22 @@
       "score": 0.7161974280320248,
       "sentence_nr": 6
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "de",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.6486932415130529,
+      "sentence_nr": 6
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "de",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.788686710424071,
+      "sentence_nr": 6
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "de",
@@ -53029,6 +61179,22 @@
       "score": 0.0,
       "sentence_nr": 6
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ja",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.39057277619092257,
+      "sentence_nr": 6
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ja",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.3934016321632531,
+      "sentence_nr": 6
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ja",
@@ -53061,6 +61227,22 @@
       "score": 0.4118595729651108,
       "sentence_nr": 6
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ja",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.7490853969372642,
+      "sentence_nr": 6
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ja",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.8062091543413888,
+      "sentence_nr": 6
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ja",
@@ -53205,6 +61387,22 @@
       "score": 0.6634154486532953,
       "sentence_nr": 7
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "en",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.343041631179768,
+      "sentence_nr": 7
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "en",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.6383283286161612,
+      "sentence_nr": 7
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "en",
@@ -53237,6 +61435,22 @@
       "score": 0.6481070648129139,
       "sentence_nr": 7
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "en",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.5702655877666989,
+      "sentence_nr": 7
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "en",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.8232854345902009,
+      "sentence_nr": 7
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "en",
@@ -53381,6 +61595,22 @@
       "score": 0.23455679137513727,
       "sentence_nr": 7
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "zh",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.0,
+      "sentence_nr": 7
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "zh",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.1775614884118737,
+      "sentence_nr": 7
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "zh",
@@ -53413,6 +61643,22 @@
       "score": 0.26662620996190534,
       "sentence_nr": 7
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "zh",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.30538115660133164,
+      "sentence_nr": 7
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "zh",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.338864446519003,
+      "sentence_nr": 7
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "zh",
@@ -53557,6 +61803,22 @@
       "score": 0.5835478395499368,
       "sentence_nr": 7
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "hi",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.11690741296020518,
+      "sentence_nr": 7
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "hi",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.44984705715342654,
+      "sentence_nr": 7
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "hi",
@@ -53589,6 +61851,22 @@
       "score": 0.21558480215297515,
       "sentence_nr": 7
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "hi",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.25639784746935274,
+      "sentence_nr": 7
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "hi",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.5733913952323451,
+      "sentence_nr": 7
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "hi",
@@ -53733,6 +62011,22 @@
       "score": 0.553839023223762,
       "sentence_nr": 7
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "es",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.24479697566202357,
+      "sentence_nr": 7
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "es",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.5213023098886357,
+      "sentence_nr": 7
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "es",
@@ -53765,6 +62059,22 @@
       "score": 0.4573855767208229,
       "sentence_nr": 7
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "es",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.45149688763848994,
+      "sentence_nr": 7
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "es",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.6166500596693574,
+      "sentence_nr": 7
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "es",
@@ -53909,6 +62219,22 @@
       "score": 0.4303467795130825,
       "sentence_nr": 7
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ar",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.2529076741385625,
+      "sentence_nr": 7
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ar",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.5430833248313275,
+      "sentence_nr": 7
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ar",
@@ -53941,6 +62267,22 @@
       "score": 0.35911678207067443,
       "sentence_nr": 7
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ar",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.4221671351559825,
+      "sentence_nr": 7
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ar",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.6302060108035411,
+      "sentence_nr": 7
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ar",
@@ -54085,6 +62427,22 @@
       "score": 0.4466759653076362,
       "sentence_nr": 7
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ur",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.28261688976967947,
+      "sentence_nr": 7
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ur",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.5075237416590358,
+      "sentence_nr": 7
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ur",
@@ -54117,6 +62475,22 @@
       "score": 0.3796663901127053,
       "sentence_nr": 7
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ur",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.3935019932504074,
+      "sentence_nr": 7
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ur",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.6071903247613194,
+      "sentence_nr": 7
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ur",
@@ -54261,6 +62635,22 @@
       "score": 0.5863866793721222,
       "sentence_nr": 7
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "fr",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.38084051173962913,
+      "sentence_nr": 7
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "fr",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.7334510090568515,
+      "sentence_nr": 7
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "fr",
@@ -54293,6 +62683,22 @@
       "score": 0.7181569025811343,
       "sentence_nr": 7
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "fr",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.6263164471220594,
+      "sentence_nr": 7
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "fr",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.852013904460107,
+      "sentence_nr": 7
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "fr",
@@ -54437,6 +62843,22 @@
       "score": 0.45303225382772006,
       "sentence_nr": 7
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "bn",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.2378706071654586,
+      "sentence_nr": 7
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "bn",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.49630155585737695,
+      "sentence_nr": 7
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "bn",
@@ -54469,6 +62891,22 @@
       "score": 0.559682285505658,
       "sentence_nr": 7
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "bn",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.36440851219076265,
+      "sentence_nr": 7
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "bn",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.6157376412237141,
+      "sentence_nr": 7
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "bn",
@@ -54613,6 +63051,22 @@
       "score": 0.45111566089364774,
       "sentence_nr": 7
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pt",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.16510868745008767,
+      "sentence_nr": 7
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pt",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.4762714594756596,
+      "sentence_nr": 7
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "pt",
@@ -54645,6 +63099,22 @@
       "score": 0.5931856951819833,
       "sentence_nr": 7
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pt",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.40696144066138723,
+      "sentence_nr": 7
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pt",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.6476222098586478,
+      "sentence_nr": 7
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "pt",
@@ -54789,6 +63259,22 @@
       "score": 0.32241875701400735,
       "sentence_nr": 7
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pa",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.2059931729749887,
+      "sentence_nr": 7
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pa",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.4491812480433597,
+      "sentence_nr": 7
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "pa",
@@ -54821,6 +63307,22 @@
       "score": 0.308102700736633,
       "sentence_nr": 7
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pa",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.35174128537520233,
+      "sentence_nr": 7
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pa",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.5232532816160403,
+      "sentence_nr": 7
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "pa",
@@ -54965,6 +63467,22 @@
       "score": 0.5698489012763526,
       "sentence_nr": 7
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ru",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.2572958792096885,
+      "sentence_nr": 7
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ru",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.5064333161464132,
+      "sentence_nr": 7
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ru",
@@ -54997,6 +63515,22 @@
       "score": 0.6060858750149657,
       "sentence_nr": 7
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ru",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.5839068685770862,
+      "sentence_nr": 7
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ru",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.7498987134962192,
+      "sentence_nr": 7
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ru",
@@ -55141,6 +63675,22 @@
       "score": 0.4170915413269471,
       "sentence_nr": 7
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "sw",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.11675747661776523,
+      "sentence_nr": 7
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "sw",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.4301556378791635,
+      "sentence_nr": 7
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "sw",
@@ -55173,6 +63723,22 @@
       "score": 0.3886262536746606,
       "sentence_nr": 7
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "sw",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.30250421218183904,
+      "sentence_nr": 7
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "sw",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.5314198518425818,
+      "sentence_nr": 7
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "sw",
@@ -55317,6 +63883,22 @@
       "score": 0.4876777357531764,
       "sentence_nr": 7
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "id",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.29019765706301537,
+      "sentence_nr": 7
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "id",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.598596485843401,
+      "sentence_nr": 7
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "id",
@@ -55349,6 +63931,22 @@
       "score": 0.6474323586139361,
       "sentence_nr": 7
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "id",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.3486251970868468,
+      "sentence_nr": 7
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "id",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.6519074063738273,
+      "sentence_nr": 7
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "id",
@@ -55493,6 +64091,22 @@
       "score": 0.6104451101668408,
       "sentence_nr": 7
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "de",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.46290680573161996,
+      "sentence_nr": 7
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "de",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.688869111662782,
+      "sentence_nr": 7
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "de",
@@ -55525,6 +64139,22 @@
       "score": 0.6707552233208028,
       "sentence_nr": 7
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "de",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.5995826896426277,
+      "sentence_nr": 7
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "de",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.7943692036315023,
+      "sentence_nr": 7
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "de",
@@ -55669,6 +64299,22 @@
       "score": 0.0680779227699037,
       "sentence_nr": 7
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ja",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.18386904980839383,
+      "sentence_nr": 7
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ja",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.3263831403119468,
+      "sentence_nr": 7
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ja",
@@ -55701,6 +64347,22 @@
       "score": 0.3579044902117876,
       "sentence_nr": 7
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ja",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.2409168844747761,
+      "sentence_nr": 7
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ja",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.3760904447135035,
+      "sentence_nr": 7
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ja",
@@ -55845,6 +64507,22 @@
       "score": 0.47660259733052845,
       "sentence_nr": 8
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "en",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.19803162353826262,
+      "sentence_nr": 8
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "en",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.40808208228398596,
+      "sentence_nr": 8
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "en",
@@ -55877,6 +64555,22 @@
       "score": 0.5719371199531044,
       "sentence_nr": 8
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "en",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.3407563025626974,
+      "sentence_nr": 8
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "en",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.6182648747061787,
+      "sentence_nr": 8
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "en",
@@ -56021,6 +64715,22 @@
       "score": 0.052821402483564636,
       "sentence_nr": 8
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "zh",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.0,
+      "sentence_nr": 8
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "zh",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.13492461680840023,
+      "sentence_nr": 8
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "zh",
@@ -56053,6 +64763,22 @@
       "score": 0.20388486867467934,
       "sentence_nr": 8
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "zh",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.1621568294816267,
+      "sentence_nr": 8
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "zh",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.2655543079691671,
+      "sentence_nr": 8
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "zh",
@@ -56197,6 +64923,22 @@
       "score": 0.4983778740634126,
       "sentence_nr": 8
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "hi",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.3916177035633811,
+      "sentence_nr": 8
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "hi",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.6068458202737596,
+      "sentence_nr": 8
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "hi",
@@ -56229,6 +64971,22 @@
       "score": 0.6068458202737596,
       "sentence_nr": 8
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "hi",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.3916177035633811,
+      "sentence_nr": 8
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "hi",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.6068458202737596,
+      "sentence_nr": 8
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "hi",
@@ -56373,6 +65131,22 @@
       "score": 0.4103553163121394,
       "sentence_nr": 8
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "es",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.0,
+      "sentence_nr": 8
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "es",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.37440084690294706,
+      "sentence_nr": 8
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "es",
@@ -56405,6 +65179,22 @@
       "score": 0.36586001924521905,
       "sentence_nr": 8
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "es",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.0,
+      "sentence_nr": 8
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "es",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.35316499124143624,
+      "sentence_nr": 8
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "es",
@@ -56549,6 +65339,22 @@
       "score": 0.4685134392551311,
       "sentence_nr": 8
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ar",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.0,
+      "sentence_nr": 8
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ar",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.3359695440470467,
+      "sentence_nr": 8
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ar",
@@ -56581,6 +65387,22 @@
       "score": 0.4378856092523028,
       "sentence_nr": 8
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ar",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.437742810290776,
+      "sentence_nr": 8
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ar",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.621154967713859,
+      "sentence_nr": 8
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ar",
@@ -56725,6 +65547,22 @@
       "score": 0.3785761836985817,
       "sentence_nr": 8
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ur",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.28336087141473976,
+      "sentence_nr": 8
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ur",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.38214734777528636,
+      "sentence_nr": 8
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ur",
@@ -56757,6 +65595,22 @@
       "score": 0.3818556455365969,
       "sentence_nr": 8
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ur",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.3266011589665084,
+      "sentence_nr": 8
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ur",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.4364077249430218,
+      "sentence_nr": 8
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ur",
@@ -56901,6 +65755,22 @@
       "score": 0.646847036932526,
       "sentence_nr": 8
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "fr",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.13857910426205777,
+      "sentence_nr": 8
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "fr",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.41268771676068494,
+      "sentence_nr": 8
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "fr",
@@ -56933,6 +65803,22 @@
       "score": 0.5619162673780028,
       "sentence_nr": 8
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "fr",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.38694317759010316,
+      "sentence_nr": 8
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "fr",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.6962124663194352,
+      "sentence_nr": 8
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "fr",
@@ -57077,6 +65963,22 @@
       "score": 0.3640275543948514,
       "sentence_nr": 8
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "bn",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.18384948243517193,
+      "sentence_nr": 8
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "bn",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.40935288248313256,
+      "sentence_nr": 8
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "bn",
@@ -57109,6 +66011,22 @@
       "score": 0.3459916112351503,
       "sentence_nr": 8
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "bn",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.31638337148949686,
+      "sentence_nr": 8
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "bn",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.5386981918746817,
+      "sentence_nr": 8
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "bn",
@@ -57253,6 +66171,22 @@
       "score": 0.6492026440953677,
       "sentence_nr": 8
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pt",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.1280220256953781,
+      "sentence_nr": 8
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pt",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.3766998614914371,
+      "sentence_nr": 8
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "pt",
@@ -57285,6 +66219,22 @@
       "score": 0.577852219465442,
       "sentence_nr": 8
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pt",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.33667089470100775,
+      "sentence_nr": 8
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pt",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.6074301230157959,
+      "sentence_nr": 8
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "pt",
@@ -57429,6 +66379,22 @@
       "score": 0.0067104198717751464,
       "sentence_nr": 8
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pa",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 1.0,
+      "sentence_nr": 8
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pa",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 1.0,
+      "sentence_nr": 8
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "pa",
@@ -57461,6 +66427,22 @@
       "score": 0.9169897590736298,
       "sentence_nr": 8
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pa",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 1.0,
+      "sentence_nr": 8
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pa",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 1.0,
+      "sentence_nr": 8
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "pa",
@@ -57605,6 +66587,22 @@
       "score": 0.4030284875466178,
       "sentence_nr": 8
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ru",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.0,
+      "sentence_nr": 8
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ru",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.38985048513980286,
+      "sentence_nr": 8
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ru",
@@ -57637,6 +66635,22 @@
       "score": 0.43120286814245795,
       "sentence_nr": 8
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ru",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.4489929043142396,
+      "sentence_nr": 8
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ru",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.6339860404289296,
+      "sentence_nr": 8
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ru",
@@ -57781,6 +66795,22 @@
       "score": 0.4536746865348185,
       "sentence_nr": 8
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "sw",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.17268932789342512,
+      "sentence_nr": 8
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "sw",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.47806194925419343,
+      "sentence_nr": 8
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "sw",
@@ -57813,6 +66843,22 @@
       "score": 0.40551062972908847,
       "sentence_nr": 8
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "sw",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.42382856047421374,
+      "sentence_nr": 8
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "sw",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.6321659688090209,
+      "sentence_nr": 8
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "sw",
@@ -57957,6 +67003,22 @@
       "score": 0.5592183664602846,
       "sentence_nr": 8
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "id",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.1528571341245854,
+      "sentence_nr": 8
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "id",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.5481360183906401,
+      "sentence_nr": 8
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "id",
@@ -57989,6 +67051,22 @@
       "score": 0.4500155517039222,
       "sentence_nr": 8
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "id",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.37219737664729546,
+      "sentence_nr": 8
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "id",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.6020217844686402,
+      "sentence_nr": 8
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "id",
@@ -58133,6 +67211,22 @@
       "score": 0.43980817368282343,
       "sentence_nr": 8
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "de",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.3323217739558646,
+      "sentence_nr": 8
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "de",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.65584145837095,
+      "sentence_nr": 8
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "de",
@@ -58165,6 +67259,22 @@
       "score": 0.5421002898382512,
       "sentence_nr": 8
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "de",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.3961867597457338,
+      "sentence_nr": 8
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "de",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.654720368848453,
+      "sentence_nr": 8
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "de",
@@ -58309,6 +67419,22 @@
       "score": 0.010715460821011002,
       "sentence_nr": 8
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ja",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.0,
+      "sentence_nr": 8
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ja",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.3235165184655995,
+      "sentence_nr": 8
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ja",
@@ -58341,6 +67467,22 @@
       "score": 0.2690830377349408,
       "sentence_nr": 8
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ja",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.22537412722674852,
+      "sentence_nr": 8
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ja",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.3609019979890711,
+      "sentence_nr": 8
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ja",
@@ -58485,6 +67627,22 @@
       "score": 0.6772940233934857,
       "sentence_nr": 9
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "en",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.34303589686600006,
+      "sentence_nr": 9
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "en",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.6476809000259773,
+      "sentence_nr": 9
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "en",
@@ -58517,6 +67675,22 @@
       "score": 0.6734455797843703,
       "sentence_nr": 9
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "en",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.4330463947479356,
+      "sentence_nr": 9
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "en",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.6198346106486459,
+      "sentence_nr": 9
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "en",
@@ -58661,6 +67835,22 @@
       "score": 0.21051700087939107,
       "sentence_nr": 9
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "zh",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.13093265020876002,
+      "sentence_nr": 9
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "zh",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.24423594551873207,
+      "sentence_nr": 9
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "zh",
@@ -58693,6 +67883,22 @@
       "score": 0.28188465375440136,
       "sentence_nr": 9
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "zh",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.17755132725434278,
+      "sentence_nr": 9
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "zh",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.27939552769052894,
+      "sentence_nr": 9
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "zh",
@@ -58837,6 +68043,22 @@
       "score": 0.3088863284587533,
       "sentence_nr": 9
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "hi",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.16561315331829,
+      "sentence_nr": 9
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "hi",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.4491486313807806,
+      "sentence_nr": 9
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "hi",
@@ -58869,6 +68091,22 @@
       "score": 0.3616605984753398,
       "sentence_nr": 9
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "hi",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.2985215837565239,
+      "sentence_nr": 9
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "hi",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.5357534811195178,
+      "sentence_nr": 9
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "hi",
@@ -59013,6 +68251,22 @@
       "score": 0.6534660189132082,
       "sentence_nr": 9
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "es",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.3118437333980883,
+      "sentence_nr": 9
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "es",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.49789195197291464,
+      "sentence_nr": 9
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "es",
@@ -59045,6 +68299,22 @@
       "score": 0.6026940597371309,
       "sentence_nr": 9
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "es",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.39653893552930464,
+      "sentence_nr": 9
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "es",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.5591087327390277,
+      "sentence_nr": 9
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "es",
@@ -59189,6 +68459,22 @@
       "score": 0.5170853673805775,
       "sentence_nr": 9
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ar",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.09950615774798431,
+      "sentence_nr": 9
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ar",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.4145753205477691,
+      "sentence_nr": 9
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ar",
@@ -59221,6 +68507,22 @@
       "score": 0.4780977009860418,
       "sentence_nr": 9
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ar",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.22546521673609302,
+      "sentence_nr": 9
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ar",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.5396972089286257,
+      "sentence_nr": 9
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ar",
@@ -59365,6 +68667,22 @@
       "score": 0.4784292149775752,
       "sentence_nr": 9
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ur",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.18555265687599404,
+      "sentence_nr": 9
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ur",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.3943451217336116,
+      "sentence_nr": 9
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ur",
@@ -59397,6 +68715,22 @@
       "score": 0.4114443619817223,
       "sentence_nr": 9
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ur",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.35541324629951093,
+      "sentence_nr": 9
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ur",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.49416627591115303,
+      "sentence_nr": 9
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ur",
@@ -59541,6 +68875,22 @@
       "score": 0.7625459507115938,
       "sentence_nr": 9
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "fr",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.38962400400495395,
+      "sentence_nr": 9
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "fr",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.6039376140178496,
+      "sentence_nr": 9
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "fr",
@@ -59573,6 +68923,22 @@
       "score": 0.58212864821275,
       "sentence_nr": 9
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "fr",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.729605098531811,
+      "sentence_nr": 9
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "fr",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.8833183865941937,
+      "sentence_nr": 9
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "fr",
@@ -59717,6 +69083,22 @@
       "score": 0.3805666011451541,
       "sentence_nr": 9
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "bn",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.0,
+      "sentence_nr": 9
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "bn",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.38511373700997104,
+      "sentence_nr": 9
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "bn",
@@ -59749,6 +69131,22 @@
       "score": 0.4677317890018283,
       "sentence_nr": 9
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "bn",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.0,
+      "sentence_nr": 9
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "bn",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.3787500122126683,
+      "sentence_nr": 9
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "bn",
@@ -59893,6 +69291,22 @@
       "score": 0.6307076431103672,
       "sentence_nr": 9
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pt",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.431319746325093,
+      "sentence_nr": 9
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pt",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.6367129659739652,
+      "sentence_nr": 9
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "pt",
@@ -59925,6 +69339,22 @@
       "score": 0.6663170490872967,
       "sentence_nr": 9
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pt",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.5471998982127312,
+      "sentence_nr": 9
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pt",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.668797436013741,
+      "sentence_nr": 9
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "pt",
@@ -60069,6 +69499,22 @@
       "score": 0.41943156806161835,
       "sentence_nr": 9
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pa",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.13950796967929138,
+      "sentence_nr": 9
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pa",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.3461520644408903,
+      "sentence_nr": 9
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "pa",
@@ -60101,6 +69547,22 @@
       "score": 0.35651447515721807,
       "sentence_nr": 9
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pa",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.3393109592089468,
+      "sentence_nr": 9
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pa",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.5094767086148101,
+      "sentence_nr": 9
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "pa",
@@ -60245,6 +69707,22 @@
       "score": 0.47811810874873667,
       "sentence_nr": 9
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ru",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.23288432092807593,
+      "sentence_nr": 9
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ru",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.571224820704715,
+      "sentence_nr": 9
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ru",
@@ -60277,6 +69755,22 @@
       "score": 0.6172751686457948,
       "sentence_nr": 9
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ru",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.39359852091634406,
+      "sentence_nr": 9
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ru",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.6265459059804013,
+      "sentence_nr": 9
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ru",
@@ -60421,6 +69915,22 @@
       "score": 0.3720123244240524,
       "sentence_nr": 9
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "sw",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.1663603558532716,
+      "sentence_nr": 9
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "sw",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.5225044219427867,
+      "sentence_nr": 9
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "sw",
@@ -60453,6 +69963,22 @@
       "score": 0.34496242859007625,
       "sentence_nr": 9
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "sw",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.429294349316905,
+      "sentence_nr": 9
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "sw",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.6516072032987875,
+      "sentence_nr": 9
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "sw",
@@ -60597,6 +70123,22 @@
       "score": 0.604856884215657,
       "sentence_nr": 9
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "id",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.24078435074822624,
+      "sentence_nr": 9
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "id",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.5511929774495076,
+      "sentence_nr": 9
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "id",
@@ -60629,6 +70171,22 @@
       "score": 0.5314568462829651,
       "sentence_nr": 9
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "id",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.31818246779032233,
+      "sentence_nr": 9
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "id",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.6229693133713873,
+      "sentence_nr": 9
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "id",
@@ -60773,6 +70331,22 @@
       "score": 0.5456825634559386,
       "sentence_nr": 9
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "de",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.18081719101162075,
+      "sentence_nr": 9
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "de",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.5207488300398541,
+      "sentence_nr": 9
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "de",
@@ -60805,6 +70379,22 @@
       "score": 0.4502822427440237,
       "sentence_nr": 9
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "de",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.4008628432878983,
+      "sentence_nr": 9
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "de",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.5958332944294813,
+      "sentence_nr": 9
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "de",
@@ -60949,6 +70539,22 @@
       "score": 0.8142101616656354,
       "sentence_nr": 9
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ja",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 0.0,
+      "sentence_nr": 9
+    },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ja",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 0.2762822897608569,
+      "sentence_nr": 9
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ja",
@@ -60981,6 +70587,22 @@
       "score": 1.0,
       "sentence_nr": 9
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ja",
+      "task": "translation_to",
+      "metric": "bleu",
+      "score": 1.0,
+      "sentence_nr": 9
+    },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ja",
+      "task": "translation_to",
+      "metric": "chrf",
+      "score": 1.0,
+      "sentence_nr": 9
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ja",
@@ -61093,6 +70715,14 @@
       "score": 1,
       "sentence_nr": 0
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "en",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 0
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "en",
@@ -61109,6 +70739,14 @@
       "score": 1,
       "sentence_nr": 0
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "en",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 0
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "en",
@@ -61181,6 +70819,14 @@
       "score": 1,
       "sentence_nr": 0
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "zh",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 0
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "zh",
@@ -61197,6 +70843,14 @@
       "score": 0,
       "sentence_nr": 0
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "zh",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 0
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "zh",
@@ -61269,6 +70923,14 @@
       "score": 1,
       "sentence_nr": 0
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "hi",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 0
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "hi",
@@ -61285,6 +70947,14 @@
       "score": 1,
       "sentence_nr": 0
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "hi",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 0
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "hi",
@@ -61357,6 +71027,14 @@
       "score": 1,
       "sentence_nr": 0
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "es",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 0
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "es",
@@ -61373,6 +71051,14 @@
       "score": 0,
       "sentence_nr": 0
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "es",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 0
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "es",
@@ -61445,6 +71131,14 @@
       "score": 0,
       "sentence_nr": 0
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ar",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 0
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ar",
@@ -61461,6 +71155,14 @@
       "score": 0,
       "sentence_nr": 0
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ar",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 0
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ar",
@@ -61533,6 +71235,14 @@
       "score": 1,
       "sentence_nr": 0
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ur",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 0
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ur",
@@ -61549,6 +71259,14 @@
       "score": 0,
       "sentence_nr": 0
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ur",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 0
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ur",
@@ -61621,6 +71339,14 @@
       "score": 1,
       "sentence_nr": 0
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "fr",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 0
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "fr",
@@ -61637,6 +71363,14 @@
       "score": 0,
       "sentence_nr": 0
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "fr",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 0
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "fr",
@@ -61709,6 +71443,14 @@
       "score": 0,
       "sentence_nr": 0
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "bn",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 0
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "bn",
@@ -61725,6 +71467,14 @@
       "score": 0,
       "sentence_nr": 0
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "bn",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 0
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "bn",
@@ -61797,6 +71547,14 @@
       "score": 1,
       "sentence_nr": 0
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pt",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 0
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "pt",
@@ -61813,6 +71571,14 @@
       "score": 0,
       "sentence_nr": 0
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pt",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 0
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "pt",
@@ -61885,6 +71651,14 @@
       "score": 0,
       "sentence_nr": 0
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pa",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 0
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "pa",
@@ -61901,6 +71675,14 @@
       "score": 1,
       "sentence_nr": 0
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pa",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 0
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "pa",
@@ -61973,6 +71755,14 @@
       "score": 1,
       "sentence_nr": 0
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ru",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 0
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ru",
@@ -61989,6 +71779,14 @@
       "score": 1,
       "sentence_nr": 0
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ru",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 0
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ru",
@@ -62061,6 +71859,14 @@
       "score": 1,
       "sentence_nr": 0
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "sw",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 0
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "sw",
@@ -62077,6 +71883,14 @@
       "score": 0,
       "sentence_nr": 0
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "sw",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 0
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "sw",
@@ -62149,6 +71963,14 @@
       "score": 1,
       "sentence_nr": 0
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "id",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 0
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "id",
@@ -62165,6 +71987,14 @@
       "score": 0,
       "sentence_nr": 0
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "id",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 0
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "id",
@@ -62237,6 +72067,14 @@
       "score": 1,
       "sentence_nr": 0
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "de",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 0
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "de",
@@ -62253,6 +72091,14 @@
       "score": 0,
       "sentence_nr": 0
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "de",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 0
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "de",
@@ -62325,6 +72171,14 @@
       "score": 1,
       "sentence_nr": 0
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ja",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 0
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ja",
@@ -62341,6 +72195,14 @@
       "score": 1,
       "sentence_nr": 0
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ja",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 0
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ja",
@@ -62413,6 +72275,14 @@
       "score": 1,
       "sentence_nr": 1
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "en",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 1
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "en",
@@ -62429,6 +72299,14 @@
       "score": 0,
       "sentence_nr": 1
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "en",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 1
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "en",
@@ -62501,6 +72379,14 @@
       "score": 1,
       "sentence_nr": 1
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "zh",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 1
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "zh",
@@ -62517,6 +72403,14 @@
       "score": 1,
       "sentence_nr": 1
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "zh",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 1
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "zh",
@@ -62589,6 +72483,14 @@
       "score": 0,
       "sentence_nr": 1
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "hi",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 1
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "hi",
@@ -62605,6 +72507,14 @@
       "score": 1,
       "sentence_nr": 1
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "hi",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 1
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "hi",
@@ -62677,6 +72587,14 @@
       "score": 1,
       "sentence_nr": 1
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "es",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 1
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "es",
@@ -62693,6 +72611,14 @@
       "score": 0,
       "sentence_nr": 1
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "es",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 1
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "es",
@@ -62765,6 +72691,14 @@
       "score": 1,
       "sentence_nr": 1
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ar",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 1
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ar",
@@ -62781,6 +72715,14 @@
       "score": 1,
       "sentence_nr": 1
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ar",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 1
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ar",
@@ -62853,6 +72795,14 @@
       "score": 1,
       "sentence_nr": 1
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ur",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 1
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ur",
@@ -62869,6 +72819,14 @@
       "score": 1,
       "sentence_nr": 1
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ur",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 1
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ur",
@@ -62941,6 +72899,14 @@
       "score": 1,
       "sentence_nr": 1
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "fr",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 1
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "fr",
@@ -62957,6 +72923,14 @@
       "score": 1,
       "sentence_nr": 1
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "fr",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 1
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "fr",
@@ -63029,6 +73003,14 @@
       "score": 0,
       "sentence_nr": 1
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "bn",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 1
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "bn",
@@ -63045,6 +73027,14 @@
       "score": 0,
       "sentence_nr": 1
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "bn",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 1
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "bn",
@@ -63117,6 +73107,14 @@
       "score": 1,
       "sentence_nr": 1
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pt",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 1
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "pt",
@@ -63133,6 +73131,14 @@
       "score": 1,
       "sentence_nr": 1
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pt",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 1
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "pt",
@@ -63205,6 +73211,14 @@
       "score": 0,
       "sentence_nr": 1
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pa",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 1
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "pa",
@@ -63221,6 +73235,14 @@
       "score": 0,
       "sentence_nr": 1
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pa",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 1
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "pa",
@@ -63293,6 +73315,14 @@
       "score": 1,
       "sentence_nr": 1
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ru",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 1
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ru",
@@ -63309,6 +73339,14 @@
       "score": 1,
       "sentence_nr": 1
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ru",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 1
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ru",
@@ -63381,6 +73419,14 @@
       "score": 1,
       "sentence_nr": 1
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "sw",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 1
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "sw",
@@ -63397,6 +73443,14 @@
       "score": 1,
       "sentence_nr": 1
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "sw",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 1
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "sw",
@@ -63469,6 +73523,14 @@
       "score": 1,
       "sentence_nr": 1
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "id",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 1
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "id",
@@ -63485,6 +73547,14 @@
       "score": 1,
       "sentence_nr": 1
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "id",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 1
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "id",
@@ -63557,6 +73627,14 @@
       "score": 1,
       "sentence_nr": 1
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "de",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 1
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "de",
@@ -63573,6 +73651,14 @@
       "score": 1,
       "sentence_nr": 1
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "de",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 1
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "de",
@@ -63645,6 +73731,14 @@
       "score": 1,
       "sentence_nr": 1
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ja",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 1
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ja",
@@ -63661,6 +73755,14 @@
       "score": 1,
       "sentence_nr": 1
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ja",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 1
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ja",
@@ -63733,6 +73835,14 @@
       "score": 1,
       "sentence_nr": 2
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "en",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 2
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "en",
@@ -63749,6 +73859,14 @@
       "score": 0,
       "sentence_nr": 2
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "en",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 2
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "en",
@@ -63821,6 +73939,14 @@
       "score": 1,
       "sentence_nr": 2
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "zh",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 2
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "zh",
@@ -63837,6 +73963,14 @@
       "score": 1,
       "sentence_nr": 2
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "zh",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 2
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "zh",
@@ -63909,6 +74043,14 @@
       "score": 1,
       "sentence_nr": 2
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "hi",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 2
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "hi",
@@ -63925,6 +74067,14 @@
       "score": 1,
       "sentence_nr": 2
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "hi",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 2
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "hi",
@@ -63997,6 +74147,14 @@
       "score": 1,
       "sentence_nr": 2
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "es",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 2
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "es",
@@ -64013,6 +74171,14 @@
       "score": 0,
       "sentence_nr": 2
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "es",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 2
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "es",
@@ -64085,6 +74251,14 @@
       "score": 1,
       "sentence_nr": 2
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ar",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 2
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ar",
@@ -64101,6 +74275,14 @@
       "score": 1,
       "sentence_nr": 2
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ar",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 2
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ar",
@@ -64173,6 +74355,14 @@
       "score": 1,
       "sentence_nr": 2
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ur",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 2
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ur",
@@ -64189,6 +74379,14 @@
       "score": 1,
       "sentence_nr": 2
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ur",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 2
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ur",
@@ -64261,6 +74459,14 @@
       "score": 1,
       "sentence_nr": 2
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "fr",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 2
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "fr",
@@ -64277,6 +74483,14 @@
       "score": 1,
       "sentence_nr": 2
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "fr",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 2
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "fr",
@@ -64349,6 +74563,14 @@
       "score": 0,
       "sentence_nr": 2
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "bn",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 2
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "bn",
@@ -64365,6 +74587,14 @@
       "score": 1,
       "sentence_nr": 2
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "bn",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 2
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "bn",
@@ -64437,6 +74667,14 @@
       "score": 1,
       "sentence_nr": 2
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pt",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 2
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "pt",
@@ -64453,6 +74691,14 @@
       "score": 1,
       "sentence_nr": 2
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pt",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 2
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "pt",
@@ -64525,6 +74771,14 @@
       "score": 0,
       "sentence_nr": 2
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pa",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 2
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "pa",
@@ -64541,6 +74795,14 @@
       "score": 1,
       "sentence_nr": 2
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pa",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 2
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "pa",
@@ -64613,6 +74875,14 @@
       "score": 1,
       "sentence_nr": 2
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ru",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 2
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ru",
@@ -64629,6 +74899,14 @@
       "score": 1,
       "sentence_nr": 2
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ru",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 2
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ru",
@@ -64701,6 +74979,14 @@
       "score": 1,
       "sentence_nr": 2
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "sw",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 2
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "sw",
@@ -64717,6 +75003,14 @@
       "score": 1,
       "sentence_nr": 2
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "sw",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 2
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "sw",
@@ -64789,6 +75083,14 @@
       "score": 1,
       "sentence_nr": 2
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "id",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 2
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "id",
@@ -64805,6 +75107,14 @@
       "score": 1,
       "sentence_nr": 2
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "id",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 2
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "id",
@@ -64877,6 +75187,14 @@
       "score": 1,
       "sentence_nr": 2
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "de",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 2
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "de",
@@ -64893,6 +75211,14 @@
       "score": 1,
       "sentence_nr": 2
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "de",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 2
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "de",
@@ -64965,6 +75291,14 @@
       "score": 1,
       "sentence_nr": 2
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ja",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 2
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ja",
@@ -64981,6 +75315,14 @@
       "score": 1,
       "sentence_nr": 2
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ja",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 2
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ja",
@@ -65053,6 +75395,14 @@
       "score": 1,
       "sentence_nr": 3
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "en",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 3
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "en",
@@ -65069,6 +75419,14 @@
       "score": 0,
       "sentence_nr": 3
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "en",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 3
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "en",
@@ -65141,6 +75499,14 @@
       "score": 1,
       "sentence_nr": 3
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "zh",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 3
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "zh",
@@ -65157,6 +75523,14 @@
       "score": 1,
       "sentence_nr": 3
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "zh",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 3
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "zh",
@@ -65229,6 +75603,14 @@
       "score": 0,
       "sentence_nr": 3
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "hi",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 3
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "hi",
@@ -65245,6 +75627,14 @@
       "score": 1,
       "sentence_nr": 3
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "hi",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 3
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "hi",
@@ -65317,6 +75707,14 @@
       "score": 1,
       "sentence_nr": 3
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "es",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 3
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "es",
@@ -65333,6 +75731,14 @@
       "score": 1,
       "sentence_nr": 3
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "es",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 3
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "es",
@@ -65405,6 +75811,14 @@
       "score": 0,
       "sentence_nr": 3
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ar",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 3
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ar",
@@ -65421,6 +75835,14 @@
       "score": 1,
       "sentence_nr": 3
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ar",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 3
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ar",
@@ -65493,6 +75915,14 @@
       "score": 0,
       "sentence_nr": 3
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ur",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 3
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ur",
@@ -65509,6 +75939,14 @@
       "score": 1,
       "sentence_nr": 3
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ur",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 3
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ur",
@@ -65581,6 +76019,14 @@
       "score": 1,
       "sentence_nr": 3
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "fr",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 3
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "fr",
@@ -65597,6 +76043,14 @@
       "score": 1,
       "sentence_nr": 3
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "fr",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 3
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "fr",
@@ -65669,6 +76123,14 @@
       "score": 0,
       "sentence_nr": 3
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "bn",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 3
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "bn",
@@ -65685,6 +76147,14 @@
       "score": 1,
       "sentence_nr": 3
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "bn",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 3
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "bn",
@@ -65757,6 +76227,14 @@
       "score": 1,
       "sentence_nr": 3
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pt",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 3
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "pt",
@@ -65773,6 +76251,14 @@
       "score": 0,
       "sentence_nr": 3
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pt",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 3
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "pt",
@@ -65845,6 +76331,14 @@
       "score": 0,
       "sentence_nr": 3
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pa",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 3
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "pa",
@@ -65861,6 +76355,14 @@
       "score": 1,
       "sentence_nr": 3
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pa",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 3
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "pa",
@@ -65933,6 +76435,14 @@
       "score": 1,
       "sentence_nr": 3
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ru",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 3
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ru",
@@ -65949,6 +76459,14 @@
       "score": 1,
       "sentence_nr": 3
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ru",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 3
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ru",
@@ -66021,6 +76539,14 @@
       "score": 1,
       "sentence_nr": 3
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "sw",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 3
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "sw",
@@ -66037,6 +76563,14 @@
       "score": 1,
       "sentence_nr": 3
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "sw",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 3
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "sw",
@@ -66109,6 +76643,14 @@
       "score": 1,
       "sentence_nr": 3
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "id",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 3
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "id",
@@ -66125,6 +76667,14 @@
       "score": 1,
       "sentence_nr": 3
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "id",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 3
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "id",
@@ -66197,6 +76747,14 @@
       "score": 1,
       "sentence_nr": 3
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "de",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 3
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "de",
@@ -66213,6 +76771,14 @@
       "score": 1,
       "sentence_nr": 3
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "de",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 3
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "de",
@@ -66285,6 +76851,14 @@
       "score": 0,
       "sentence_nr": 3
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ja",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 3
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ja",
@@ -66301,6 +76875,14 @@
       "score": 0,
       "sentence_nr": 3
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ja",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 3
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ja",
@@ -66373,6 +76955,14 @@
       "score": 1,
       "sentence_nr": 4
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "en",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 4
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "en",
@@ -66389,6 +76979,14 @@
       "score": 0,
       "sentence_nr": 4
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "en",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 4
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "en",
@@ -66461,6 +77059,14 @@
       "score": 1,
       "sentence_nr": 4
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "zh",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 4
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "zh",
@@ -66477,6 +77083,14 @@
       "score": 0,
       "sentence_nr": 4
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "zh",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 4
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "zh",
@@ -66549,6 +77163,14 @@
       "score": 1,
       "sentence_nr": 4
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "hi",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 4
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "hi",
@@ -66565,6 +77187,14 @@
       "score": 0,
       "sentence_nr": 4
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "hi",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 4
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "hi",
@@ -66637,6 +77267,14 @@
       "score": 1,
       "sentence_nr": 4
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "es",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 4
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "es",
@@ -66653,6 +77291,14 @@
       "score": 0,
       "sentence_nr": 4
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "es",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 4
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "es",
@@ -66725,6 +77371,14 @@
       "score": 1,
       "sentence_nr": 4
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ar",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 4
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ar",
@@ -66741,6 +77395,14 @@
       "score": 0,
       "sentence_nr": 4
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ar",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 4
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ar",
@@ -66813,6 +77475,14 @@
       "score": 1,
       "sentence_nr": 4
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ur",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 4
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ur",
@@ -66829,6 +77499,14 @@
       "score": 1,
       "sentence_nr": 4
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ur",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 4
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ur",
@@ -66901,6 +77579,14 @@
       "score": 0,
       "sentence_nr": 4
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "fr",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 4
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "fr",
@@ -66917,6 +77603,14 @@
       "score": 0,
       "sentence_nr": 4
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "fr",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 4
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "fr",
@@ -66989,6 +77683,14 @@
       "score": 1,
       "sentence_nr": 4
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "bn",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 4
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "bn",
@@ -67005,6 +77707,14 @@
       "score": 0,
       "sentence_nr": 4
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "bn",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 4
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "bn",
@@ -67077,6 +77787,14 @@
       "score": 1,
       "sentence_nr": 4
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pt",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 4
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "pt",
@@ -67093,6 +77811,14 @@
       "score": 0,
       "sentence_nr": 4
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pt",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 4
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "pt",
@@ -67165,6 +77891,14 @@
       "score": 1,
       "sentence_nr": 4
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pa",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 4
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "pa",
@@ -67181,6 +77915,14 @@
       "score": 0,
       "sentence_nr": 4
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pa",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 4
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "pa",
@@ -67253,6 +77995,14 @@
       "score": 1,
       "sentence_nr": 4
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ru",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 4
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ru",
@@ -67269,6 +78019,14 @@
       "score": 0,
       "sentence_nr": 4
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ru",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 4
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ru",
@@ -67341,6 +78099,14 @@
       "score": 0,
       "sentence_nr": 4
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "sw",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 4
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "sw",
@@ -67357,6 +78123,14 @@
       "score": 1,
       "sentence_nr": 4
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "sw",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 4
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "sw",
@@ -67429,6 +78203,14 @@
       "score": 1,
       "sentence_nr": 4
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "id",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 4
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "id",
@@ -67445,6 +78227,14 @@
       "score": 1,
       "sentence_nr": 4
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "id",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 4
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "id",
@@ -67517,6 +78307,14 @@
       "score": 1,
       "sentence_nr": 4
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "de",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 4
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "de",
@@ -67533,6 +78331,14 @@
       "score": 1,
       "sentence_nr": 4
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "de",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 4
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "de",
@@ -67605,6 +78411,14 @@
       "score": 1,
       "sentence_nr": 4
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ja",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 4
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ja",
@@ -67621,6 +78435,14 @@
       "score": 0,
       "sentence_nr": 4
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ja",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 4
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ja",
@@ -67693,6 +78515,14 @@
       "score": 1,
       "sentence_nr": 5
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "en",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 5
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "en",
@@ -67709,6 +78539,14 @@
       "score": 0,
       "sentence_nr": 5
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "en",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 5
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "en",
@@ -67781,6 +78619,14 @@
       "score": 1,
       "sentence_nr": 5
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "zh",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 5
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "zh",
@@ -67797,6 +78643,14 @@
       "score": 1,
       "sentence_nr": 5
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "zh",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 5
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "zh",
@@ -67869,6 +78723,14 @@
       "score": 1,
       "sentence_nr": 5
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "hi",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 5
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "hi",
@@ -67885,6 +78747,14 @@
       "score": 1,
       "sentence_nr": 5
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "hi",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 5
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "hi",
@@ -67957,6 +78827,14 @@
       "score": 1,
       "sentence_nr": 5
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "es",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 5
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "es",
@@ -67973,6 +78851,14 @@
       "score": 1,
       "sentence_nr": 5
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "es",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 5
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "es",
@@ -68045,6 +78931,14 @@
       "score": 1,
       "sentence_nr": 5
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ar",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 5
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ar",
@@ -68061,6 +78955,14 @@
       "score": 1,
       "sentence_nr": 5
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ar",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 5
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ar",
@@ -68133,6 +79035,14 @@
       "score": 1,
       "sentence_nr": 5
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ur",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 5
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ur",
@@ -68149,6 +79059,14 @@
       "score": 1,
       "sentence_nr": 5
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ur",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 5
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ur",
@@ -68221,6 +79139,14 @@
       "score": 1,
       "sentence_nr": 5
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "fr",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 5
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "fr",
@@ -68237,6 +79163,14 @@
       "score": 1,
       "sentence_nr": 5
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "fr",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 5
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "fr",
@@ -68309,6 +79243,14 @@
       "score": 0,
       "sentence_nr": 5
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "bn",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 5
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "bn",
@@ -68325,6 +79267,14 @@
       "score": 1,
       "sentence_nr": 5
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "bn",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 5
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "bn",
@@ -68397,6 +79347,14 @@
       "score": 1,
       "sentence_nr": 5
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pt",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 5
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "pt",
@@ -68413,6 +79371,14 @@
       "score": 1,
       "sentence_nr": 5
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pt",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 5
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "pt",
@@ -68485,6 +79451,14 @@
       "score": 0,
       "sentence_nr": 5
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pa",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 5
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "pa",
@@ -68501,6 +79475,14 @@
       "score": 1,
       "sentence_nr": 5
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pa",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 5
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "pa",
@@ -68573,6 +79555,14 @@
       "score": 1,
       "sentence_nr": 5
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ru",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 5
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ru",
@@ -68589,6 +79579,14 @@
       "score": 1,
       "sentence_nr": 5
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ru",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 5
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ru",
@@ -68661,6 +79659,14 @@
       "score": 1,
       "sentence_nr": 5
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "sw",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 5
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "sw",
@@ -68677,6 +79683,14 @@
       "score": 1,
       "sentence_nr": 5
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "sw",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 5
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "sw",
@@ -68749,6 +79763,14 @@
       "score": 1,
       "sentence_nr": 5
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "id",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 5
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "id",
@@ -68765,6 +79787,14 @@
       "score": 1,
       "sentence_nr": 5
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "id",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 5
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "id",
@@ -68837,6 +79867,14 @@
       "score": 1,
       "sentence_nr": 5
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "de",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 5
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "de",
@@ -68853,6 +79891,14 @@
       "score": 1,
       "sentence_nr": 5
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "de",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 5
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "de",
@@ -68925,6 +79971,14 @@
       "score": 1,
       "sentence_nr": 5
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ja",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 5
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ja",
@@ -68941,6 +79995,14 @@
       "score": 0,
       "sentence_nr": 5
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ja",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 5
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ja",
@@ -69013,6 +80075,14 @@
       "score": 1,
       "sentence_nr": 6
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "en",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 6
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "en",
@@ -69029,6 +80099,14 @@
       "score": 0,
       "sentence_nr": 6
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "en",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 6
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "en",
@@ -69101,6 +80179,14 @@
       "score": 0,
       "sentence_nr": 6
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "zh",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 6
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "zh",
@@ -69117,6 +80203,14 @@
       "score": 1,
       "sentence_nr": 6
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "zh",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 6
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "zh",
@@ -69189,6 +80283,14 @@
       "score": 0,
       "sentence_nr": 6
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "hi",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 6
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "hi",
@@ -69205,6 +80307,14 @@
       "score": 0,
       "sentence_nr": 6
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "hi",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 6
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "hi",
@@ -69277,6 +80387,14 @@
       "score": 1,
       "sentence_nr": 6
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "es",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 6
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "es",
@@ -69293,6 +80411,14 @@
       "score": 0,
       "sentence_nr": 6
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "es",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 6
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "es",
@@ -69365,6 +80491,14 @@
       "score": 0,
       "sentence_nr": 6
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ar",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 6
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ar",
@@ -69381,6 +80515,14 @@
       "score": 0,
       "sentence_nr": 6
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ar",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 6
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ar",
@@ -69453,6 +80595,14 @@
       "score": 0,
       "sentence_nr": 6
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ur",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 6
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ur",
@@ -69469,6 +80619,14 @@
       "score": 0,
       "sentence_nr": 6
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ur",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 6
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ur",
@@ -69541,6 +80699,14 @@
       "score": 1,
       "sentence_nr": 6
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "fr",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 6
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "fr",
@@ -69557,6 +80723,14 @@
       "score": 0,
       "sentence_nr": 6
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "fr",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 6
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "fr",
@@ -69629,6 +80803,14 @@
       "score": 0,
       "sentence_nr": 6
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "bn",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 6
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "bn",
@@ -69645,6 +80827,14 @@
       "score": 1,
       "sentence_nr": 6
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "bn",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 6
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "bn",
@@ -69717,6 +80907,14 @@
       "score": 1,
       "sentence_nr": 6
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pt",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 6
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "pt",
@@ -69733,6 +80931,14 @@
       "score": 0,
       "sentence_nr": 6
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pt",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 6
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "pt",
@@ -69805,6 +81011,14 @@
       "score": 0,
       "sentence_nr": 6
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pa",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 6
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "pa",
@@ -69821,6 +81035,14 @@
       "score": 0,
       "sentence_nr": 6
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pa",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 6
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "pa",
@@ -69893,6 +81115,14 @@
       "score": 0,
       "sentence_nr": 6
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ru",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 6
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ru",
@@ -69909,6 +81139,14 @@
       "score": 0,
       "sentence_nr": 6
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ru",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 6
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ru",
@@ -69981,6 +81219,14 @@
       "score": 0,
       "sentence_nr": 6
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "sw",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 6
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "sw",
@@ -69997,6 +81243,14 @@
       "score": 1,
       "sentence_nr": 6
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "sw",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 6
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "sw",
@@ -70069,6 +81323,14 @@
       "score": 0,
       "sentence_nr": 6
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "id",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 6
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "id",
@@ -70085,6 +81347,14 @@
       "score": 0,
       "sentence_nr": 6
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "id",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 6
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "id",
@@ -70157,6 +81427,14 @@
       "score": 1,
       "sentence_nr": 6
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "de",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 6
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "de",
@@ -70173,6 +81451,14 @@
       "score": 0,
       "sentence_nr": 6
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "de",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 6
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "de",
@@ -70245,6 +81531,14 @@
       "score": 0,
       "sentence_nr": 6
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ja",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 6
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ja",
@@ -70261,6 +81555,14 @@
       "score": 0,
       "sentence_nr": 6
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ja",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 6
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ja",
@@ -70333,6 +81635,14 @@
       "score": 1,
       "sentence_nr": 7
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "en",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 7
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "en",
@@ -70349,6 +81659,14 @@
       "score": 0,
       "sentence_nr": 7
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "en",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 7
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "en",
@@ -70421,6 +81739,14 @@
       "score": 0,
       "sentence_nr": 7
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "zh",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 7
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "zh",
@@ -70437,6 +81763,14 @@
       "score": 0,
       "sentence_nr": 7
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "zh",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 7
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "zh",
@@ -70509,6 +81843,14 @@
       "score": 0,
       "sentence_nr": 7
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "hi",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 7
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "hi",
@@ -70525,6 +81867,14 @@
       "score": 0,
       "sentence_nr": 7
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "hi",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 7
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "hi",
@@ -70597,6 +81947,14 @@
       "score": 0,
       "sentence_nr": 7
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "es",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 7
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "es",
@@ -70613,6 +81971,14 @@
       "score": 0,
       "sentence_nr": 7
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "es",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 7
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "es",
@@ -70685,6 +82051,14 @@
       "score": 0,
       "sentence_nr": 7
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ar",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 7
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ar",
@@ -70701,6 +82075,14 @@
       "score": 0,
       "sentence_nr": 7
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ar",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 7
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ar",
@@ -70773,6 +82155,14 @@
       "score": 0,
       "sentence_nr": 7
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ur",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 7
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ur",
@@ -70789,6 +82179,14 @@
       "score": 0,
       "sentence_nr": 7
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ur",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 7
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ur",
@@ -70861,6 +82259,14 @@
       "score": 0,
       "sentence_nr": 7
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "fr",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 7
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "fr",
@@ -70877,6 +82283,14 @@
       "score": 0,
       "sentence_nr": 7
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "fr",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 7
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "fr",
@@ -70949,6 +82363,14 @@
       "score": 0,
       "sentence_nr": 7
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "bn",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 7
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "bn",
@@ -70965,6 +82387,14 @@
       "score": 0,
       "sentence_nr": 7
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "bn",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 7
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "bn",
@@ -71037,6 +82467,14 @@
       "score": 0,
       "sentence_nr": 7
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pt",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 7
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "pt",
@@ -71053,6 +82491,14 @@
       "score": 0,
       "sentence_nr": 7
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pt",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 7
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "pt",
@@ -71125,6 +82571,14 @@
       "score": 0,
       "sentence_nr": 7
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pa",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 7
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "pa",
@@ -71141,6 +82595,14 @@
       "score": 0,
       "sentence_nr": 7
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pa",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 7
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "pa",
@@ -71213,6 +82675,14 @@
       "score": 0,
       "sentence_nr": 7
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ru",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 7
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ru",
@@ -71229,6 +82699,14 @@
       "score": 0,
       "sentence_nr": 7
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ru",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 7
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ru",
@@ -71301,6 +82779,14 @@
       "score": 0,
       "sentence_nr": 7
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "sw",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 7
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "sw",
@@ -71317,6 +82803,14 @@
       "score": 0,
       "sentence_nr": 7
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "sw",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 7
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "sw",
@@ -71389,6 +82883,14 @@
       "score": 0,
       "sentence_nr": 7
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "id",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 7
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "id",
@@ -71405,6 +82907,14 @@
       "score": 0,
       "sentence_nr": 7
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "id",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 7
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "id",
@@ -71477,6 +82987,14 @@
       "score": 0,
       "sentence_nr": 7
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "de",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 7
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "de",
@@ -71493,6 +83011,14 @@
       "score": 0,
       "sentence_nr": 7
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "de",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 7
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "de",
@@ -71565,6 +83091,14 @@
       "score": 0,
       "sentence_nr": 7
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ja",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 7
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ja",
@@ -71581,6 +83115,14 @@
       "score": 0,
       "sentence_nr": 7
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ja",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 7
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ja",
@@ -71653,6 +83195,14 @@
       "score": 1,
       "sentence_nr": 8
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "en",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 8
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "en",
@@ -71669,6 +83219,14 @@
       "score": 1,
       "sentence_nr": 8
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "en",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 8
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "en",
@@ -71741,6 +83299,14 @@
       "score": 1,
       "sentence_nr": 8
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "zh",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 8
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "zh",
@@ -71757,6 +83323,14 @@
       "score": 0,
       "sentence_nr": 8
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "zh",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 8
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "zh",
@@ -71829,6 +83403,14 @@
       "score": 0,
       "sentence_nr": 8
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "hi",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 8
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "hi",
@@ -71845,6 +83427,14 @@
       "score": 0,
       "sentence_nr": 8
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "hi",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 8
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "hi",
@@ -71917,6 +83507,14 @@
       "score": 1,
       "sentence_nr": 8
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "es",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 8
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "es",
@@ -71933,6 +83531,14 @@
       "score": 0,
       "sentence_nr": 8
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "es",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 8
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "es",
@@ -72005,6 +83611,14 @@
       "score": 0,
       "sentence_nr": 8
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ar",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 8
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ar",
@@ -72021,6 +83635,14 @@
       "score": 0,
       "sentence_nr": 8
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ar",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 8
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ar",
@@ -72093,6 +83715,14 @@
       "score": 0,
       "sentence_nr": 8
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ur",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 8
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ur",
@@ -72109,6 +83739,14 @@
       "score": 1,
       "sentence_nr": 8
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ur",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 8
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ur",
@@ -72181,6 +83819,14 @@
       "score": 1,
       "sentence_nr": 8
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "fr",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 8
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "fr",
@@ -72197,6 +83843,14 @@
       "score": 0,
       "sentence_nr": 8
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "fr",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 8
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "fr",
@@ -72269,6 +83923,14 @@
       "score": 0,
       "sentence_nr": 8
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "bn",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 8
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "bn",
@@ -72285,6 +83947,14 @@
       "score": 1,
       "sentence_nr": 8
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "bn",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 8
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "bn",
@@ -72357,6 +84027,14 @@
       "score": 1,
       "sentence_nr": 8
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pt",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 8
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "pt",
@@ -72373,6 +84051,14 @@
       "score": 0,
       "sentence_nr": 8
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pt",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 8
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "pt",
@@ -72445,6 +84131,14 @@
       "score": 0,
       "sentence_nr": 8
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pa",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 8
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "pa",
@@ -72461,6 +84155,14 @@
       "score": 0,
       "sentence_nr": 8
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pa",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 8
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "pa",
@@ -72533,6 +84235,14 @@
       "score": 1,
       "sentence_nr": 8
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ru",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 8
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ru",
@@ -72549,6 +84259,14 @@
       "score": 1,
       "sentence_nr": 8
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ru",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 8
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ru",
@@ -72621,6 +84339,14 @@
       "score": 1,
       "sentence_nr": 8
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "sw",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 8
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "sw",
@@ -72637,6 +84363,14 @@
       "score": 1,
       "sentence_nr": 8
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "sw",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 8
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "sw",
@@ -72709,6 +84443,14 @@
       "score": 1,
       "sentence_nr": 8
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "id",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 8
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "id",
@@ -72725,6 +84467,14 @@
       "score": 1,
       "sentence_nr": 8
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "id",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 8
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "id",
@@ -72797,6 +84547,14 @@
       "score": 1,
       "sentence_nr": 8
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "de",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 8
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "de",
@@ -72813,6 +84571,14 @@
       "score": 1,
       "sentence_nr": 8
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "de",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 8
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "de",
@@ -72885,6 +84651,14 @@
       "score": 1,
       "sentence_nr": 8
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ja",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 8
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ja",
@@ -72901,6 +84675,14 @@
       "score": 1,
       "sentence_nr": 8
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ja",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 8
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ja",
@@ -72973,6 +84755,14 @@
       "score": 0,
       "sentence_nr": 9
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "en",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 9
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "en",
@@ -72989,6 +84779,14 @@
       "score": 0,
       "sentence_nr": 9
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "en",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 9
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "en",
@@ -73061,6 +84859,14 @@
       "score": 0,
       "sentence_nr": 9
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "zh",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 9
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "zh",
@@ -73077,6 +84883,14 @@
       "score": 0,
       "sentence_nr": 9
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "zh",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 9
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "zh",
@@ -73149,6 +84963,14 @@
       "score": 0,
       "sentence_nr": 9
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "hi",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 9
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "hi",
@@ -73165,6 +84987,14 @@
       "score": 0,
       "sentence_nr": 9
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "hi",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 9
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "hi",
@@ -73237,6 +85067,14 @@
       "score": 0,
       "sentence_nr": 9
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "es",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 9
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "es",
@@ -73253,6 +85091,14 @@
       "score": 0,
       "sentence_nr": 9
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "es",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 9
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "es",
@@ -73325,6 +85171,14 @@
       "score": 0,
       "sentence_nr": 9
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ar",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 9
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ar",
@@ -73341,6 +85195,14 @@
       "score": 0,
       "sentence_nr": 9
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ar",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 9
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ar",
@@ -73413,6 +85275,14 @@
       "score": 0,
       "sentence_nr": 9
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ur",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 9
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ur",
@@ -73429,6 +85299,14 @@
       "score": 0,
       "sentence_nr": 9
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ur",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 9
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ur",
@@ -73501,6 +85379,14 @@
       "score": 0,
       "sentence_nr": 9
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "fr",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 9
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "fr",
@@ -73517,6 +85403,14 @@
       "score": 0,
       "sentence_nr": 9
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "fr",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 9
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "fr",
@@ -73589,6 +85483,14 @@
       "score": 0,
       "sentence_nr": 9
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "bn",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 9
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "bn",
@@ -73605,6 +85507,14 @@
       "score": 0,
       "sentence_nr": 9
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "bn",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 9
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "bn",
@@ -73677,6 +85587,14 @@
       "score": 0,
       "sentence_nr": 9
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pt",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 9
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "pt",
@@ -73693,6 +85611,14 @@
       "score": 0,
       "sentence_nr": 9
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pt",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 9
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "pt",
@@ -73765,6 +85691,14 @@
       "score": 0,
       "sentence_nr": 9
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pa",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 9
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "pa",
@@ -73781,6 +85715,14 @@
       "score": 0,
       "sentence_nr": 9
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pa",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 9
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "pa",
@@ -73853,6 +85795,14 @@
       "score": 0,
       "sentence_nr": 9
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ru",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 9
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ru",
@@ -73869,6 +85819,14 @@
       "score": 0,
       "sentence_nr": 9
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ru",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 9
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ru",
@@ -73941,6 +85899,14 @@
       "score": 0,
       "sentence_nr": 9
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "sw",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 9
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "sw",
@@ -73957,6 +85923,14 @@
       "score": 0,
       "sentence_nr": 9
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "sw",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 9
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "sw",
@@ -74029,6 +86003,14 @@
       "score": 0,
       "sentence_nr": 9
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "id",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 9
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "id",
@@ -74045,6 +86027,14 @@
       "score": 0,
       "sentence_nr": 9
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "id",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 9
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "id",
@@ -74117,6 +86107,14 @@
       "score": 0,
       "sentence_nr": 9
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "de",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 9
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "de",
@@ -74133,6 +86131,14 @@
       "score": 0,
       "sentence_nr": 9
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "de",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 9
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "de",
@@ -74205,6 +86211,14 @@
       "score": 0,
       "sentence_nr": 9
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ja",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 9
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ja",
@@ -74221,6 +86235,14 @@
       "score": 0,
       "sentence_nr": 9
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ja",
+      "task": "classification",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 9
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ja",
@@ -74293,6 +86315,14 @@
       "score": 1,
       "sentence_nr": 0
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "en",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 0
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "en",
@@ -74309,6 +86339,14 @@
       "score": 1,
       "sentence_nr": 0
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "en",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 0
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "en",
@@ -74381,6 +86419,14 @@
       "score": 1,
       "sentence_nr": 0
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "zh",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 0
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "zh",
@@ -74397,6 +86443,14 @@
       "score": 1,
       "sentence_nr": 0
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "zh",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 0
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "zh",
@@ -74469,6 +86523,14 @@
       "score": 1,
       "sentence_nr": 0
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "hi",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 0
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "hi",
@@ -74485,6 +86547,14 @@
       "score": 1,
       "sentence_nr": 0
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "hi",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 0
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "hi",
@@ -74557,6 +86627,14 @@
       "score": 1,
       "sentence_nr": 0
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "es",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 0
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "es",
@@ -74573,6 +86651,14 @@
       "score": 1,
       "sentence_nr": 0
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "es",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 0
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "es",
@@ -74645,6 +86731,14 @@
       "score": 1,
       "sentence_nr": 0
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ar",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 0
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ar",
@@ -74661,6 +86755,14 @@
       "score": 1,
       "sentence_nr": 0
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ar",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 0
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ar",
@@ -74733,6 +86835,14 @@
       "score": 0,
       "sentence_nr": 0
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "fr",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 0
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "fr",
@@ -74749,6 +86859,14 @@
       "score": 0,
       "sentence_nr": 0
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "fr",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 0
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "fr",
@@ -74821,6 +86939,14 @@
       "score": 1,
       "sentence_nr": 0
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "bn",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 0
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "bn",
@@ -74837,6 +86963,14 @@
       "score": 1,
       "sentence_nr": 0
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "bn",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 0
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "bn",
@@ -74909,6 +87043,14 @@
       "score": 1,
       "sentence_nr": 0
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pt",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 0
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "pt",
@@ -74925,6 +87067,14 @@
       "score": 1,
       "sentence_nr": 0
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pt",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 0
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "pt",
@@ -74997,6 +87147,14 @@
       "score": 1,
       "sentence_nr": 0
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ru",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 0
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ru",
@@ -75013,6 +87171,14 @@
       "score": 1,
       "sentence_nr": 0
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ru",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 0
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ru",
@@ -75085,6 +87251,14 @@
       "score": 1,
       "sentence_nr": 0
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "sw",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 0
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "sw",
@@ -75101,6 +87275,14 @@
       "score": 1,
       "sentence_nr": 0
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "sw",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 0
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "sw",
@@ -75173,6 +87355,14 @@
       "score": 1,
       "sentence_nr": 0
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "id",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 0
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "id",
@@ -75189,6 +87379,14 @@
       "score": 1,
       "sentence_nr": 0
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "id",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 0
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "id",
@@ -75261,6 +87459,14 @@
       "score": 1,
       "sentence_nr": 0
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "de",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 0
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "de",
@@ -75277,6 +87483,14 @@
       "score": 1,
       "sentence_nr": 0
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "de",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 0
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "de",
@@ -75349,6 +87563,14 @@
       "score": 1,
       "sentence_nr": 0
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ja",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 0
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ja",
@@ -75365,6 +87587,14 @@
       "score": 1,
       "sentence_nr": 0
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ja",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 0
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ja",
@@ -75437,6 +87667,14 @@
       "score": 1,
       "sentence_nr": 1
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "en",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 1
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "en",
@@ -75453,6 +87691,14 @@
       "score": 1,
       "sentence_nr": 1
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "en",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 1
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "en",
@@ -75525,6 +87771,14 @@
       "score": 1,
       "sentence_nr": 1
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "zh",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 1
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "zh",
@@ -75541,6 +87795,14 @@
       "score": 1,
       "sentence_nr": 1
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "zh",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 1
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "zh",
@@ -75613,6 +87875,14 @@
       "score": 1,
       "sentence_nr": 1
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "hi",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 1
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "hi",
@@ -75629,6 +87899,14 @@
       "score": 1,
       "sentence_nr": 1
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "hi",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 1
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "hi",
@@ -75701,6 +87979,14 @@
       "score": 1,
       "sentence_nr": 1
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "es",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 1
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "es",
@@ -75717,6 +88003,14 @@
       "score": 1,
       "sentence_nr": 1
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "es",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 1
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "es",
@@ -75789,6 +88083,14 @@
       "score": 1,
       "sentence_nr": 1
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ar",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 1
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ar",
@@ -75805,6 +88107,14 @@
       "score": 1,
       "sentence_nr": 1
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ar",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 1
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ar",
@@ -75877,6 +88187,14 @@
       "score": 1,
       "sentence_nr": 1
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "fr",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 1
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "fr",
@@ -75893,6 +88211,14 @@
       "score": 1,
       "sentence_nr": 1
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "fr",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 1
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "fr",
@@ -75965,6 +88291,14 @@
       "score": 1,
       "sentence_nr": 1
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "bn",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 1
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "bn",
@@ -75981,6 +88315,14 @@
       "score": 1,
       "sentence_nr": 1
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "bn",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 1
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "bn",
@@ -76053,6 +88395,14 @@
       "score": 1,
       "sentence_nr": 1
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pt",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 1
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "pt",
@@ -76069,6 +88419,14 @@
       "score": 1,
       "sentence_nr": 1
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pt",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 1
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "pt",
@@ -76141,6 +88499,14 @@
       "score": 1,
       "sentence_nr": 1
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ru",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 1
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ru",
@@ -76157,6 +88523,14 @@
       "score": 1,
       "sentence_nr": 1
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ru",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 1
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ru",
@@ -76229,6 +88603,14 @@
       "score": 1,
       "sentence_nr": 1
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "sw",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 1
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "sw",
@@ -76245,6 +88627,14 @@
       "score": 1,
       "sentence_nr": 1
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "sw",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 1
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "sw",
@@ -76317,6 +88707,14 @@
       "score": 1,
       "sentence_nr": 1
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "id",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 1
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "id",
@@ -76333,6 +88731,14 @@
       "score": 1,
       "sentence_nr": 1
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "id",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 1
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "id",
@@ -76405,6 +88811,14 @@
       "score": 1,
       "sentence_nr": 1
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "de",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 1
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "de",
@@ -76421,6 +88835,14 @@
       "score": 1,
       "sentence_nr": 1
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "de",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 1
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "de",
@@ -76493,6 +88915,14 @@
       "score": 1,
       "sentence_nr": 1
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ja",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 1
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ja",
@@ -76509,6 +88939,14 @@
       "score": 1,
       "sentence_nr": 1
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ja",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 1
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ja",
@@ -76581,6 +89019,14 @@
       "score": 1,
       "sentence_nr": 2
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "en",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 2
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "en",
@@ -76597,6 +89043,14 @@
       "score": 1,
       "sentence_nr": 2
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "en",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 2
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "en",
@@ -76669,6 +89123,14 @@
       "score": 1,
       "sentence_nr": 2
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "zh",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 2
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "zh",
@@ -76685,6 +89147,14 @@
       "score": 1,
       "sentence_nr": 2
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "zh",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 2
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "zh",
@@ -76757,6 +89227,14 @@
       "score": 1,
       "sentence_nr": 2
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "hi",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 2
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "hi",
@@ -76773,6 +89251,14 @@
       "score": 1,
       "sentence_nr": 2
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "hi",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 2
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "hi",
@@ -76845,6 +89331,14 @@
       "score": 1,
       "sentence_nr": 2
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "es",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 2
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "es",
@@ -76861,6 +89355,14 @@
       "score": 1,
       "sentence_nr": 2
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "es",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 2
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "es",
@@ -76933,6 +89435,14 @@
       "score": 1,
       "sentence_nr": 2
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ar",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 2
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ar",
@@ -76949,6 +89459,14 @@
       "score": 1,
       "sentence_nr": 2
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ar",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 2
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ar",
@@ -77021,6 +89539,14 @@
       "score": 1,
       "sentence_nr": 2
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "fr",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 2
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "fr",
@@ -77037,6 +89563,14 @@
       "score": 1,
       "sentence_nr": 2
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "fr",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 2
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "fr",
@@ -77109,6 +89643,14 @@
       "score": 1,
       "sentence_nr": 2
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "bn",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 2
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "bn",
@@ -77125,6 +89667,14 @@
       "score": 1,
       "sentence_nr": 2
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "bn",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 2
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "bn",
@@ -77197,6 +89747,14 @@
       "score": 1,
       "sentence_nr": 2
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pt",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 2
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "pt",
@@ -77213,6 +89771,14 @@
       "score": 1,
       "sentence_nr": 2
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pt",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 2
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "pt",
@@ -77285,6 +89851,14 @@
       "score": 1,
       "sentence_nr": 2
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ru",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 2
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ru",
@@ -77301,6 +89875,14 @@
       "score": 1,
       "sentence_nr": 2
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ru",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 2
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ru",
@@ -77373,6 +89955,14 @@
       "score": 1,
       "sentence_nr": 2
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "sw",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 2
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "sw",
@@ -77389,6 +89979,14 @@
       "score": 0,
       "sentence_nr": 2
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "sw",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 2
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "sw",
@@ -77461,6 +90059,14 @@
       "score": 1,
       "sentence_nr": 2
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "id",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 2
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "id",
@@ -77477,6 +90083,14 @@
       "score": 1,
       "sentence_nr": 2
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "id",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 2
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "id",
@@ -77549,6 +90163,14 @@
       "score": 1,
       "sentence_nr": 2
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "de",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 2
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "de",
@@ -77565,6 +90187,14 @@
       "score": 1,
       "sentence_nr": 2
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "de",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 2
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "de",
@@ -77637,6 +90267,14 @@
       "score": 1,
       "sentence_nr": 2
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ja",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 2
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ja",
@@ -77653,6 +90291,14 @@
       "score": 1,
       "sentence_nr": 2
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ja",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 2
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ja",
@@ -77725,6 +90371,14 @@
       "score": 0,
       "sentence_nr": 3
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "en",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 3
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "en",
@@ -77741,6 +90395,14 @@
       "score": 0,
       "sentence_nr": 3
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "en",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 3
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "en",
@@ -77813,6 +90475,14 @@
       "score": 1,
       "sentence_nr": 3
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "zh",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 3
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "zh",
@@ -77829,6 +90499,14 @@
       "score": 1,
       "sentence_nr": 3
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "zh",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 3
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "zh",
@@ -77901,6 +90579,14 @@
       "score": 1,
       "sentence_nr": 3
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "hi",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 3
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "hi",
@@ -77917,6 +90603,14 @@
       "score": 0,
       "sentence_nr": 3
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "hi",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 3
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "hi",
@@ -77989,6 +90683,14 @@
       "score": 1,
       "sentence_nr": 3
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "es",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 3
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "es",
@@ -78005,6 +90707,14 @@
       "score": 1,
       "sentence_nr": 3
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "es",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 3
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "es",
@@ -78077,6 +90787,14 @@
       "score": 1,
       "sentence_nr": 3
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ar",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 3
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ar",
@@ -78093,6 +90811,14 @@
       "score": 1,
       "sentence_nr": 3
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ar",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 3
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ar",
@@ -78165,6 +90891,14 @@
       "score": 0,
       "sentence_nr": 3
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "fr",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 3
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "fr",
@@ -78181,6 +90915,14 @@
       "score": 0,
       "sentence_nr": 3
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "fr",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 3
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "fr",
@@ -78253,6 +90995,14 @@
       "score": 1,
       "sentence_nr": 3
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "bn",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 3
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "bn",
@@ -78269,6 +91019,14 @@
       "score": 0,
       "sentence_nr": 3
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "bn",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 3
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "bn",
@@ -78341,6 +91099,14 @@
       "score": 1,
       "sentence_nr": 3
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pt",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 3
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "pt",
@@ -78357,6 +91123,14 @@
       "score": 1,
       "sentence_nr": 3
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pt",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 3
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "pt",
@@ -78429,6 +91203,14 @@
       "score": 1,
       "sentence_nr": 3
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ru",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 3
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ru",
@@ -78445,6 +91227,14 @@
       "score": 1,
       "sentence_nr": 3
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ru",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 3
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ru",
@@ -78517,6 +91307,14 @@
       "score": 1,
       "sentence_nr": 3
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "sw",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 3
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "sw",
@@ -78533,6 +91331,14 @@
       "score": 1,
       "sentence_nr": 3
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "sw",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 3
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "sw",
@@ -78605,6 +91411,14 @@
       "score": 1,
       "sentence_nr": 3
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "id",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 3
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "id",
@@ -78621,6 +91435,14 @@
       "score": 1,
       "sentence_nr": 3
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "id",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 3
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "id",
@@ -78693,6 +91515,14 @@
       "score": 1,
       "sentence_nr": 3
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "de",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 3
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "de",
@@ -78709,6 +91539,14 @@
       "score": 1,
       "sentence_nr": 3
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "de",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 3
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "de",
@@ -78781,6 +91619,14 @@
       "score": 0,
       "sentence_nr": 3
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ja",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 3
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ja",
@@ -78797,6 +91643,14 @@
       "score": 1,
       "sentence_nr": 3
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ja",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 3
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ja",
@@ -78869,6 +91723,14 @@
       "score": 1,
       "sentence_nr": 4
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "en",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 4
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "en",
@@ -78885,6 +91747,14 @@
       "score": 1,
       "sentence_nr": 4
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "en",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 4
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "en",
@@ -78957,6 +91827,14 @@
       "score": 1,
       "sentence_nr": 4
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "zh",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 4
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "zh",
@@ -78973,6 +91851,14 @@
       "score": 1,
       "sentence_nr": 4
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "zh",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 4
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "zh",
@@ -79045,6 +91931,14 @@
       "score": 1,
       "sentence_nr": 4
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "hi",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 4
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "hi",
@@ -79061,6 +91955,14 @@
       "score": 1,
       "sentence_nr": 4
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "hi",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 4
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "hi",
@@ -79133,6 +92035,14 @@
       "score": 1,
       "sentence_nr": 4
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "es",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 4
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "es",
@@ -79149,6 +92059,14 @@
       "score": 1,
       "sentence_nr": 4
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "es",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 4
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "es",
@@ -79221,6 +92139,14 @@
       "score": 1,
       "sentence_nr": 4
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ar",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 4
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ar",
@@ -79237,6 +92163,14 @@
       "score": 0,
       "sentence_nr": 4
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ar",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 4
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ar",
@@ -79309,6 +92243,14 @@
       "score": 1,
       "sentence_nr": 4
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "fr",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 4
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "fr",
@@ -79325,6 +92267,14 @@
       "score": 1,
       "sentence_nr": 4
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "fr",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 4
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "fr",
@@ -79397,6 +92347,14 @@
       "score": 0,
       "sentence_nr": 4
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "bn",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 4
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "bn",
@@ -79413,6 +92371,14 @@
       "score": 0,
       "sentence_nr": 4
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "bn",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 4
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "bn",
@@ -79485,6 +92451,14 @@
       "score": 1,
       "sentence_nr": 4
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pt",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 4
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "pt",
@@ -79501,6 +92475,14 @@
       "score": 1,
       "sentence_nr": 4
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pt",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 4
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "pt",
@@ -79573,6 +92555,14 @@
       "score": 0,
       "sentence_nr": 4
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ru",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 4
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ru",
@@ -79589,6 +92579,14 @@
       "score": 1,
       "sentence_nr": 4
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ru",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 4
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ru",
@@ -79661,6 +92659,14 @@
       "score": 1,
       "sentence_nr": 4
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "sw",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 4
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "sw",
@@ -79677,6 +92683,14 @@
       "score": 1,
       "sentence_nr": 4
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "sw",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 4
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "sw",
@@ -79749,6 +92763,14 @@
       "score": 1,
       "sentence_nr": 4
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "id",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 4
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "id",
@@ -79765,6 +92787,14 @@
       "score": 1,
       "sentence_nr": 4
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "id",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 4
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "id",
@@ -79837,6 +92867,14 @@
       "score": 0,
       "sentence_nr": 4
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "de",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 4
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "de",
@@ -79853,6 +92891,14 @@
       "score": 1,
       "sentence_nr": 4
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "de",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 4
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "de",
@@ -79925,6 +92971,14 @@
       "score": 1,
       "sentence_nr": 4
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ja",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 4
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ja",
@@ -79941,6 +92995,14 @@
       "score": 1,
       "sentence_nr": 4
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ja",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 4
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ja",
@@ -80013,6 +93075,14 @@
       "score": 1,
       "sentence_nr": 5
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "en",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 5
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "en",
@@ -80029,6 +93099,14 @@
       "score": 1,
       "sentence_nr": 5
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "en",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 5
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "en",
@@ -80101,6 +93179,14 @@
       "score": 1,
       "sentence_nr": 5
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "zh",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 5
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "zh",
@@ -80117,6 +93203,14 @@
       "score": 1,
       "sentence_nr": 5
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "zh",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 5
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "zh",
@@ -80189,6 +93283,14 @@
       "score": 1,
       "sentence_nr": 5
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "hi",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 5
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "hi",
@@ -80205,6 +93307,14 @@
       "score": 1,
       "sentence_nr": 5
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "hi",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 5
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "hi",
@@ -80277,6 +93387,14 @@
       "score": 1,
       "sentence_nr": 5
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "es",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 5
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "es",
@@ -80293,6 +93411,14 @@
       "score": 1,
       "sentence_nr": 5
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "es",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 5
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "es",
@@ -80365,6 +93491,14 @@
       "score": 1,
       "sentence_nr": 5
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ar",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 5
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ar",
@@ -80381,6 +93515,14 @@
       "score": 1,
       "sentence_nr": 5
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ar",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 5
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ar",
@@ -80453,6 +93595,14 @@
       "score": 1,
       "sentence_nr": 5
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "fr",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 5
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "fr",
@@ -80469,6 +93619,14 @@
       "score": 1,
       "sentence_nr": 5
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "fr",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 5
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "fr",
@@ -80541,6 +93699,14 @@
       "score": 0,
       "sentence_nr": 5
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "bn",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 5
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "bn",
@@ -80557,6 +93723,14 @@
       "score": 1,
       "sentence_nr": 5
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "bn",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 5
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "bn",
@@ -80629,6 +93803,14 @@
       "score": 1,
       "sentence_nr": 5
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pt",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 5
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "pt",
@@ -80645,6 +93827,14 @@
       "score": 1,
       "sentence_nr": 5
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pt",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 5
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "pt",
@@ -80717,6 +93907,14 @@
       "score": 1,
       "sentence_nr": 5
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ru",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 5
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ru",
@@ -80733,6 +93931,14 @@
       "score": 1,
       "sentence_nr": 5
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ru",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 5
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ru",
@@ -80805,6 +94011,14 @@
       "score": 0,
       "sentence_nr": 5
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "sw",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 5
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "sw",
@@ -80821,6 +94035,14 @@
       "score": 1,
       "sentence_nr": 5
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "sw",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 5
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "sw",
@@ -80893,6 +94115,14 @@
       "score": 1,
       "sentence_nr": 5
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "id",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 5
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "id",
@@ -80909,6 +94139,14 @@
       "score": 1,
       "sentence_nr": 5
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "id",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 5
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "id",
@@ -80981,6 +94219,14 @@
       "score": 1,
       "sentence_nr": 5
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "de",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 5
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "de",
@@ -80997,6 +94243,14 @@
       "score": 1,
       "sentence_nr": 5
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "de",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 5
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "de",
@@ -81069,6 +94323,14 @@
       "score": 1,
       "sentence_nr": 5
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ja",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 5
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ja",
@@ -81085,6 +94347,14 @@
       "score": 1,
       "sentence_nr": 5
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ja",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 5
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ja",
@@ -81157,6 +94427,14 @@
       "score": 0,
       "sentence_nr": 6
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "en",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 6
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "en",
@@ -81173,6 +94451,14 @@
       "score": 0,
       "sentence_nr": 6
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "en",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 6
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "en",
@@ -81245,6 +94531,14 @@
       "score": 0,
       "sentence_nr": 6
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "zh",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 6
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "zh",
@@ -81261,6 +94555,14 @@
       "score": 0,
       "sentence_nr": 6
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "zh",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 6
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "zh",
@@ -81333,6 +94635,14 @@
       "score": 1,
       "sentence_nr": 6
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "hi",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 6
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "hi",
@@ -81349,6 +94659,14 @@
       "score": 0,
       "sentence_nr": 6
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "hi",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 6
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "hi",
@@ -81421,6 +94739,14 @@
       "score": 1,
       "sentence_nr": 6
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "es",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 6
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "es",
@@ -81437,6 +94763,14 @@
       "score": 0,
       "sentence_nr": 6
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "es",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 6
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "es",
@@ -81509,6 +94843,14 @@
       "score": 1,
       "sentence_nr": 6
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ar",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 6
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ar",
@@ -81525,6 +94867,14 @@
       "score": 0,
       "sentence_nr": 6
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ar",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 6
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ar",
@@ -81597,6 +94947,14 @@
       "score": 1,
       "sentence_nr": 6
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "fr",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 6
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "fr",
@@ -81613,6 +94971,14 @@
       "score": 0,
       "sentence_nr": 6
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "fr",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 6
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "fr",
@@ -81685,6 +95051,14 @@
       "score": 1,
       "sentence_nr": 6
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "bn",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 6
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "bn",
@@ -81701,6 +95075,14 @@
       "score": 0,
       "sentence_nr": 6
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "bn",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 6
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "bn",
@@ -81773,6 +95155,14 @@
       "score": 1,
       "sentence_nr": 6
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pt",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 6
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "pt",
@@ -81789,6 +95179,14 @@
       "score": 0,
       "sentence_nr": 6
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pt",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 6
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "pt",
@@ -81861,6 +95259,14 @@
       "score": 0,
       "sentence_nr": 6
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ru",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 6
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ru",
@@ -81877,6 +95283,14 @@
       "score": 0,
       "sentence_nr": 6
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ru",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 6
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ru",
@@ -81949,6 +95363,14 @@
       "score": 0,
       "sentence_nr": 6
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "sw",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 6
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "sw",
@@ -81965,6 +95387,14 @@
       "score": 0,
       "sentence_nr": 6
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "sw",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 6
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "sw",
@@ -82037,6 +95467,14 @@
       "score": 1,
       "sentence_nr": 6
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "id",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 6
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "id",
@@ -82053,6 +95491,14 @@
       "score": 0,
       "sentence_nr": 6
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "id",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 6
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "id",
@@ -82125,6 +95571,14 @@
       "score": 0,
       "sentence_nr": 6
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "de",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 6
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "de",
@@ -82141,6 +95595,14 @@
       "score": 0,
       "sentence_nr": 6
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "de",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 6
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "de",
@@ -82213,6 +95675,14 @@
       "score": 0,
       "sentence_nr": 6
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ja",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 6
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ja",
@@ -82229,6 +95699,14 @@
       "score": 0,
       "sentence_nr": 6
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ja",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 6
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ja",
@@ -82301,6 +95779,14 @@
       "score": 1,
       "sentence_nr": 7
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "en",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 7
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "en",
@@ -82317,6 +95803,14 @@
       "score": 0,
       "sentence_nr": 7
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "en",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 7
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "en",
@@ -82389,6 +95883,14 @@
       "score": 0,
       "sentence_nr": 7
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "zh",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 7
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "zh",
@@ -82405,6 +95907,14 @@
       "score": 0,
       "sentence_nr": 7
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "zh",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 7
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "zh",
@@ -82477,6 +95987,14 @@
       "score": 0,
       "sentence_nr": 7
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "hi",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 7
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "hi",
@@ -82493,6 +96011,14 @@
       "score": 1,
       "sentence_nr": 7
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "hi",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 7
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "hi",
@@ -82565,6 +96091,14 @@
       "score": 0,
       "sentence_nr": 7
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "es",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 7
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "es",
@@ -82581,6 +96115,14 @@
       "score": 0,
       "sentence_nr": 7
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "es",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 7
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "es",
@@ -82653,6 +96195,14 @@
       "score": 1,
       "sentence_nr": 7
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ar",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 7
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ar",
@@ -82669,6 +96219,14 @@
       "score": 1,
       "sentence_nr": 7
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ar",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 7
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ar",
@@ -82741,6 +96299,14 @@
       "score": 1,
       "sentence_nr": 7
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "fr",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 7
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "fr",
@@ -82757,6 +96323,14 @@
       "score": 0,
       "sentence_nr": 7
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "fr",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 7
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "fr",
@@ -82829,6 +96403,14 @@
       "score": 0,
       "sentence_nr": 7
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "bn",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 7
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "bn",
@@ -82845,6 +96427,14 @@
       "score": 0,
       "sentence_nr": 7
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "bn",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 7
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "bn",
@@ -82917,6 +96507,14 @@
       "score": 0,
       "sentence_nr": 7
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pt",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 7
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "pt",
@@ -82933,6 +96531,14 @@
       "score": 0,
       "sentence_nr": 7
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pt",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 7
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "pt",
@@ -83005,6 +96611,14 @@
       "score": 0,
       "sentence_nr": 7
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ru",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 7
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ru",
@@ -83021,6 +96635,14 @@
       "score": 1,
       "sentence_nr": 7
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ru",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 7
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ru",
@@ -83093,6 +96715,14 @@
       "score": 0,
       "sentence_nr": 7
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "sw",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 7
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "sw",
@@ -83109,6 +96739,14 @@
       "score": 0,
       "sentence_nr": 7
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "sw",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 7
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "sw",
@@ -83181,6 +96819,14 @@
       "score": 1,
       "sentence_nr": 7
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "id",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 7
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "id",
@@ -83197,6 +96843,14 @@
       "score": 0,
       "sentence_nr": 7
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "id",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 7
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "id",
@@ -83269,6 +96923,14 @@
       "score": 1,
       "sentence_nr": 7
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "de",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 7
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "de",
@@ -83285,6 +96947,14 @@
       "score": 0,
       "sentence_nr": 7
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "de",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 7
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "de",
@@ -83357,6 +97027,14 @@
       "score": 0,
       "sentence_nr": 7
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ja",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 7
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ja",
@@ -83373,6 +97051,14 @@
       "score": 0,
       "sentence_nr": 7
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ja",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 7
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ja",
@@ -83445,6 +97131,14 @@
       "score": 1,
       "sentence_nr": 8
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "en",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 8
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "en",
@@ -83461,6 +97155,14 @@
       "score": 1,
       "sentence_nr": 8
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "en",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 8
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "en",
@@ -83533,6 +97235,14 @@
       "score": 1,
       "sentence_nr": 8
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "zh",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 8
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "zh",
@@ -83549,6 +97259,14 @@
       "score": 1,
       "sentence_nr": 8
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "zh",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 8
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "zh",
@@ -83621,6 +97339,14 @@
       "score": 1,
       "sentence_nr": 8
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "hi",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 8
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "hi",
@@ -83637,6 +97363,14 @@
       "score": 1,
       "sentence_nr": 8
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "hi",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 8
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "hi",
@@ -83709,6 +97443,14 @@
       "score": 1,
       "sentence_nr": 8
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "es",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 8
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "es",
@@ -83725,6 +97467,14 @@
       "score": 1,
       "sentence_nr": 8
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "es",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 8
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "es",
@@ -83797,6 +97547,14 @@
       "score": 1,
       "sentence_nr": 8
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ar",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 8
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ar",
@@ -83813,6 +97571,14 @@
       "score": 1,
       "sentence_nr": 8
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ar",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 8
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ar",
@@ -83885,6 +97651,14 @@
       "score": 1,
       "sentence_nr": 8
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "fr",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 8
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "fr",
@@ -83901,6 +97675,14 @@
       "score": 1,
       "sentence_nr": 8
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "fr",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 8
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "fr",
@@ -83973,6 +97755,14 @@
       "score": 1,
       "sentence_nr": 8
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "bn",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 8
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "bn",
@@ -83989,6 +97779,14 @@
       "score": 1,
       "sentence_nr": 8
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "bn",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 8
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "bn",
@@ -84061,6 +97859,14 @@
       "score": 1,
       "sentence_nr": 8
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pt",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 8
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "pt",
@@ -84077,6 +97883,14 @@
       "score": 1,
       "sentence_nr": 8
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pt",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 8
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "pt",
@@ -84149,6 +97963,14 @@
       "score": 1,
       "sentence_nr": 8
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ru",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 8
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ru",
@@ -84165,6 +97987,14 @@
       "score": 1,
       "sentence_nr": 8
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ru",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 8
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ru",
@@ -84237,6 +98067,14 @@
       "score": 0,
       "sentence_nr": 8
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "sw",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 8
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "sw",
@@ -84253,6 +98091,14 @@
       "score": 0,
       "sentence_nr": 8
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "sw",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 8
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "sw",
@@ -84325,6 +98171,14 @@
       "score": 1,
       "sentence_nr": 8
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "id",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 8
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "id",
@@ -84341,6 +98195,14 @@
       "score": 1,
       "sentence_nr": 8
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "id",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 8
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "id",
@@ -84413,6 +98275,14 @@
       "score": 1,
       "sentence_nr": 8
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "de",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 8
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "de",
@@ -84429,6 +98299,14 @@
       "score": 1,
       "sentence_nr": 8
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "de",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 8
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "de",
@@ -84501,6 +98379,14 @@
       "score": 1,
       "sentence_nr": 8
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ja",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 8
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ja",
@@ -84517,6 +98403,14 @@
       "score": 1,
       "sentence_nr": 8
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ja",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 8
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ja",
@@ -84589,6 +98483,14 @@
       "score": 1,
       "sentence_nr": 9
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "en",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 9
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "en",
@@ -84605,6 +98507,14 @@
       "score": 1,
       "sentence_nr": 9
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "en",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 9
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "en",
@@ -84677,6 +98587,14 @@
       "score": 1,
       "sentence_nr": 9
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "zh",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 9
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "zh",
@@ -84693,6 +98611,14 @@
       "score": 1,
       "sentence_nr": 9
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "zh",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 9
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "zh",
@@ -84765,6 +98691,14 @@
       "score": 1,
       "sentence_nr": 9
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "hi",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 9
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "hi",
@@ -84781,6 +98715,14 @@
       "score": 1,
       "sentence_nr": 9
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "hi",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 9
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "hi",
@@ -84853,6 +98795,14 @@
       "score": 1,
       "sentence_nr": 9
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "es",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 9
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "es",
@@ -84869,6 +98819,14 @@
       "score": 1,
       "sentence_nr": 9
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "es",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 9
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "es",
@@ -84941,6 +98899,14 @@
       "score": 1,
       "sentence_nr": 9
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ar",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 9
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ar",
@@ -84957,6 +98923,14 @@
       "score": 1,
       "sentence_nr": 9
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ar",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 9
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ar",
@@ -85029,6 +99003,14 @@
       "score": 1,
       "sentence_nr": 9
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "fr",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 9
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "fr",
@@ -85045,6 +99027,14 @@
       "score": 1,
       "sentence_nr": 9
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "fr",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 9
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "fr",
@@ -85117,6 +99107,14 @@
       "score": 1,
       "sentence_nr": 9
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "bn",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 9
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "bn",
@@ -85133,6 +99131,14 @@
       "score": 0,
       "sentence_nr": 9
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "bn",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 9
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "bn",
@@ -85205,6 +99211,14 @@
       "score": 1,
       "sentence_nr": 9
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "pt",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 9
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "pt",
@@ -85221,6 +99235,14 @@
       "score": 1,
       "sentence_nr": 9
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "pt",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 9
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "pt",
@@ -85293,6 +99315,14 @@
       "score": 1,
       "sentence_nr": 9
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ru",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 9
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ru",
@@ -85309,6 +99339,14 @@
       "score": 1,
       "sentence_nr": 9
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ru",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 9
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ru",
@@ -85381,6 +99419,14 @@
       "score": 1,
       "sentence_nr": 9
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "sw",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 9
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "sw",
@@ -85397,6 +99443,14 @@
       "score": 0,
       "sentence_nr": 9
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "sw",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 9
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "sw",
@@ -85469,6 +99523,14 @@
       "score": 1,
       "sentence_nr": 9
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "id",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 9
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "id",
@@ -85485,6 +99547,14 @@
       "score": 1,
       "sentence_nr": 9
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "id",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 9
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "id",
@@ -85557,6 +99627,14 @@
       "score": 1,
       "sentence_nr": 9
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "de",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 9
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "de",
@@ -85573,6 +99651,14 @@
       "score": 1,
       "sentence_nr": 9
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "de",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 9
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "de",
@@ -85645,6 +99731,14 @@
       "score": 1,
       "sentence_nr": 9
     },
+    {
+      "model": "openai/gpt-4.1-nano",
+      "bcp_47": "ja",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 0,
+      "sentence_nr": 9
+    },
     {
       "model": "openai/gpt-4o-mini",
       "bcp_47": "ja",
@@ -85661,6 +99755,14 @@
       "score": 1,
       "sentence_nr": 9
     },
+    {
+      "model": "google/gemini-2.5-flash-preview",
+      "bcp_47": "ja",
+      "task": "mmlu",
+      "metric": "accuracy",
+      "score": 1,
+      "sentence_nr": 9
+    },
     {
       "model": "google/gemma-3-27b-it",
       "bcp_47": "ja",