hf-transformers-bot commited on
Commit
fd041d1
·
verified ·
1 Parent(s): 44e9aa1

Upload 2025-06-24/runs/171-15840041567/ci_results_run_models_gpu/new_failures_against_15840018506.json with huggingface_hub

Browse files
2025-06-24/runs/171-15840041567/ci_results_run_models_gpu/new_failures_against_15840018506.json ADDED
@@ -0,0 +1,1394 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "fsdp": {
3
+ "single-gpu": [],
4
+ "multi-gpu": [
5
+ "tests/fsdp/test_fsdp.py::TrainerIntegrationFSDP::test_basic_run_full_shard_bf16",
6
+ "tests/fsdp/test_fsdp.py::TrainerIntegrationFSDP::test_basic_run_full_shard_fp16",
7
+ "tests/fsdp/test_fsdp.py::TrainerIntegrationFSDP::test_basic_run_shard_grad_op_bf16",
8
+ "tests/fsdp/test_fsdp.py::TrainerIntegrationFSDP::test_basic_run_shard_grad_op_fp16",
9
+ "tests/fsdp/test_fsdp.py::TrainerIntegrationFSDP::test_basic_run_with_cpu_offload_0_bf16",
10
+ "tests/fsdp/test_fsdp.py::TrainerIntegrationFSDP::test_basic_run_with_cpu_offload_1_fp16",
11
+ "tests/fsdp/test_fsdp.py::TrainerIntegrationFSDP::test_basic_run_with_gradient_accumulation_full_shard_bf16",
12
+ "tests/fsdp/test_fsdp.py::TrainerIntegrationFSDP::test_basic_run_with_gradient_accumulation_full_shard_fp16",
13
+ "tests/fsdp/test_fsdp.py::TrainerIntegrationFSDP::test_basic_run_with_gradient_accumulation_shard_grad_op_bf16",
14
+ "tests/fsdp/test_fsdp.py::TrainerIntegrationFSDP::test_basic_run_with_gradient_accumulation_shard_grad_op_fp16",
15
+ "tests/fsdp/test_fsdp.py::TrainerIntegrationFSDP::test_training_and_can_resume_normally_FULL_STATE_DICT",
16
+ "tests/fsdp/test_fsdp.py::TrainerIntegrationFSDP::test_training_and_can_resume_normally_SHARDED_STATE_DICT"
17
+ ]
18
+ },
19
+ "generation": {
20
+ "single-gpu": [
21
+ "tests/generation/test_utils.py::GenerationIntegrationTests::test_assisted_decoding_num_assistant_tokens_heuristic_schedule",
22
+ "tests/generation/test_utils.py::GenerationIntegrationTests::test_cache_device_map_with_vision_layer_device_map",
23
+ "tests/generation/test_utils.py::GenerationIntegrationTests::test_constrained_beam_search_mixed",
24
+ "tests/generation/test_utils.py::GenerationIntegrationTests::test_constrained_beam_search_mixed_mixin",
25
+ "tests/generation/test_utils.py::GenerationIntegrationTests::test_custom_logits_processor",
26
+ "tests/generation/test_utils.py::GenerationIntegrationTests::test_custom_stopping_criteria_overload_error",
27
+ "tests/generation/test_utils.py::GenerationIntegrationTests::test_default_max_length_warning",
28
+ "tests/generation/test_utils.py::GenerationIntegrationTests::test_eos_token_id_int_and_list_top_k_top_sampling",
29
+ "tests/generation/test_utils.py::GenerationIntegrationTests::test_generate_compile_fullgraph_tiny",
30
+ "tests/generation/test_utils.py::GenerationIntegrationTests::test_generate_encoder_outputs_attention_mask",
31
+ "tests/generation/test_utils.py::GenerationIntegrationTests::test_generate_input_features_as_encoder_kwarg",
32
+ "tests/generation/test_utils.py::GenerationIntegrationTests::test_generate_input_ids_as_encoder_kwarg",
33
+ "tests/generation/test_utils.py::GenerationIntegrationTests::test_min_length_if_input_embeds",
34
+ "tests/generation/test_utils.py::GenerationIntegrationTests::test_model_kwarg_assisted_decoding_decoder_only",
35
+ "tests/generation/test_utils.py::GenerationIntegrationTests::test_model_kwarg_encoder_signature_filtering",
36
+ "tests/generation/test_utils.py::GenerationIntegrationTests::test_prepare_inputs_for_generation_decoder_llm",
37
+ "tests/generation/test_utils.py::GenerationIntegrationTests::test_stop_sequence_stopping_criteria"
38
+ ],
39
+ "multi-gpu": [
40
+ "tests/generation/test_utils.py::GenerationIntegrationTests::test_assisted_decoding_num_assistant_tokens_heuristic_schedule",
41
+ "tests/generation/test_utils.py::GenerationIntegrationTests::test_cache_device_map_with_vision_layer_device_map",
42
+ "tests/generation/test_utils.py::GenerationIntegrationTests::test_constrained_beam_search_mixed",
43
+ "tests/generation/test_utils.py::GenerationIntegrationTests::test_constrained_beam_search_mixed_mixin",
44
+ "tests/generation/test_utils.py::GenerationIntegrationTests::test_custom_logits_processor",
45
+ "tests/generation/test_utils.py::GenerationIntegrationTests::test_custom_stopping_criteria_overload_error",
46
+ "tests/generation/test_utils.py::GenerationIntegrationTests::test_default_max_length_warning",
47
+ "tests/generation/test_utils.py::GenerationIntegrationTests::test_eos_token_id_int_and_list_top_k_top_sampling",
48
+ "tests/generation/test_utils.py::GenerationIntegrationTests::test_generate_compile_fullgraph_tiny",
49
+ "tests/generation/test_utils.py::GenerationIntegrationTests::test_generate_encoder_outputs_attention_mask",
50
+ "tests/generation/test_utils.py::GenerationIntegrationTests::test_generate_input_features_as_encoder_kwarg",
51
+ "tests/generation/test_utils.py::GenerationIntegrationTests::test_generate_input_ids_as_encoder_kwarg",
52
+ "tests/generation/test_utils.py::GenerationIntegrationTests::test_min_length_if_input_embeds",
53
+ "tests/generation/test_utils.py::GenerationIntegrationTests::test_model_kwarg_assisted_decoding_decoder_only",
54
+ "tests/generation/test_utils.py::GenerationIntegrationTests::test_model_kwarg_encoder_signature_filtering",
55
+ "tests/generation/test_utils.py::GenerationIntegrationTests::test_prepare_inputs_for_generation_decoder_llm",
56
+ "tests/generation/test_utils.py::GenerationIntegrationTests::test_stop_sequence_stopping_criteria",
57
+ "tests/generation/test_utils.py::GenerationIntegrationTests::test_generate_multi_accelerator_causal_mask"
58
+ ]
59
+ },
60
+ "aya_vision": {
61
+ "single-gpu": [
62
+ "tests/models/aya_vision/test_modeling_aya_vision.py::AyaVisionIntegrationTest::test_small_model_integration_generate_text_only"
63
+ ],
64
+ "multi-gpu": [
65
+ "tests/models/aya_vision/test_modeling_aya_vision.py::AyaVisionIntegrationTest::test_small_model_integration_generate_text_only"
66
+ ]
67
+ },
68
+ "bamba": {
69
+ "single-gpu": [
70
+ "tests/models/bamba/test_modeling_bamba.py::BambaModelIntegrationTest::test_simple_batched_generate_with_padding",
71
+ "tests/models/bamba/test_modeling_bamba.py::BambaModelIntegrationTest::test_simple_generate"
72
+ ],
73
+ "multi-gpu": [
74
+ "tests/models/bamba/test_modeling_bamba.py::BambaModelIntegrationTest::test_simple_batched_generate_with_padding",
75
+ "tests/models/bamba/test_modeling_bamba.py::BambaModelIntegrationTest::test_simple_generate"
76
+ ]
77
+ },
78
+ "bark": {
79
+ "single-gpu": [
80
+ "tests/models/bark/test_modeling_bark.py::BarkModelIntegrationTests::test_generate_batching",
81
+ "tests/models/bark/test_modeling_bark.py::BarkModelIntegrationTests::test_generate_end_to_end_with_args"
82
+ ],
83
+ "multi-gpu": [
84
+ "tests/models/bark/test_modeling_bark.py::BarkModelIntegrationTests::test_generate_batching",
85
+ "tests/models/bark/test_modeling_bark.py::BarkModelIntegrationTests::test_generate_end_to_end_with_args"
86
+ ]
87
+ },
88
+ "beit": {
89
+ "single-gpu": [
90
+ "tests/models/beit/test_modeling_beit.py::BeitModelIntegrationTest::test_inference_semantic_segmentation",
91
+ "tests/models/beit/test_modeling_beit.py::BeitModelIntegrationTest::test_post_processing_semantic_segmentation"
92
+ ],
93
+ "multi-gpu": [
94
+ "tests/models/beit/test_modeling_beit.py::BeitModelIntegrationTest::test_inference_semantic_segmentation",
95
+ "tests/models/beit/test_modeling_beit.py::BeitModelIntegrationTest::test_post_processing_semantic_segmentation"
96
+ ]
97
+ },
98
+ "bit": {
99
+ "single-gpu": [
100
+ "tests/models/bit/test_modeling_bit.py::BitModelIntegrationTest::test_inference_image_classification_head"
101
+ ],
102
+ "multi-gpu": [
103
+ "tests/models/bit/test_modeling_bit.py::BitModelIntegrationTest::test_inference_image_classification_head"
104
+ ]
105
+ },
106
+ "bitnet": {
107
+ "single-gpu": [
108
+ "tests/models/bitnet/test_modeling_bitnet.py::BitNetIntegrationTest::test_model_logits"
109
+ ],
110
+ "multi-gpu": [
111
+ "tests/models/bitnet/test_modeling_bitnet.py::BitNetIntegrationTest::test_model_logits"
112
+ ]
113
+ },
114
+ "blip": {
115
+ "single-gpu": [],
116
+ "multi-gpu": [
117
+ "tests/models/blip/test_modeling_blip.py::BlipVQAModelTest::test_model_parallelism",
118
+ "tests/models/blip/test_modeling_blip.py::BlipTextRetrievalModelTest::test_model_parallelism",
119
+ "tests/models/blip/test_modeling_blip.py::BlipTextImageModelTest::test_model_parallelism"
120
+ ]
121
+ },
122
+ "blip_2": {
123
+ "single-gpu": [
124
+ "tests/models/blip_2/test_modeling_blip_2.py::Blip2ForConditionalGenerationDecoderOnlyTest::test_eager_matches_sdpa_generate",
125
+ "tests/models/blip_2/test_modeling_blip_2.py::Blip2ModelTest::test_eager_matches_sdpa_generate"
126
+ ],
127
+ "multi-gpu": [
128
+ "tests/models/blip_2/test_modeling_blip_2.py::Blip2ForConditionalGenerationDecoderOnlyTest::test_eager_matches_sdpa_generate",
129
+ "tests/models/blip_2/test_modeling_blip_2.py::Blip2ModelTest::test_eager_matches_sdpa_generate",
130
+ "tests/models/blip_2/test_modeling_blip_2.py::Blip2ForConditionalGenerationDecoderOnlyTest::test_model_parallelism",
131
+ "tests/models/blip_2/test_modeling_blip_2.py::Blip2ModelIntegrationTest::test_inference_opt_multi_accelerator",
132
+ "tests/models/blip_2/test_modeling_blip_2.py::Blip2ModelIntegrationTest::test_inference_t5_multi_accelerator"
133
+ ]
134
+ },
135
+ "bridgetower": {
136
+ "single-gpu": [
137
+ "tests/models/bridgetower/test_modeling_bridgetower.py::BridgeTowerModelIntegrationTest::test_masked_language_modeling"
138
+ ],
139
+ "multi-gpu": [
140
+ "tests/models/bridgetower/test_modeling_bridgetower.py::BridgeTowerModelIntegrationTest::test_masked_language_modeling"
141
+ ]
142
+ },
143
+ "chameleon": {
144
+ "single-gpu": [],
145
+ "multi-gpu": [
146
+ "tests/models/chameleon/test_modeling_chameleon.py::ChameleonVision2SeqModelTest::test_model_parallelism"
147
+ ]
148
+ },
149
+ "clvp": {
150
+ "single-gpu": [
151
+ "tests/models/clvp/test_modeling_clvp.py::ClvpIntegrationTest::test_full_model_integration"
152
+ ],
153
+ "multi-gpu": [
154
+ "tests/models/clvp/test_modeling_clvp.py::ClvpIntegrationTest::test_full_model_integration"
155
+ ]
156
+ },
157
+ "codegen": {
158
+ "single-gpu": [
159
+ "tests/models/codegen/test_modeling_codegen.py::CodeGenModelTest::test_batch_generation",
160
+ "tests/models/codegen/test_modeling_codegen.py::CodeGenModelLanguageGenerationTest::test_codegen_sample",
161
+ "tests/models/codegen/test_modeling_codegen.py::CodeGenModelLanguageGenerationTest::test_lm_generate_codegen"
162
+ ],
163
+ "multi-gpu": [
164
+ "tests/models/codegen/test_modeling_codegen.py::CodeGenModelTest::test_batch_generation",
165
+ "tests/models/codegen/test_modeling_codegen.py::CodeGenModelLanguageGenerationTest::test_codegen_sample",
166
+ "tests/models/codegen/test_modeling_codegen.py::CodeGenModelLanguageGenerationTest::test_lm_generate_codegen"
167
+ ]
168
+ },
169
+ "cohere2": {
170
+ "single-gpu": [
171
+ "tests/models/cohere2/test_modeling_cohere2.py::Cohere2IntegrationTest::test_export_static_cache",
172
+ "tests/models/cohere2/test_modeling_cohere2.py::Cohere2IntegrationTest::test_model_fp16",
173
+ "tests/models/cohere2/test_modeling_cohere2.py::Cohere2IntegrationTest::test_model_pipeline_bf16"
174
+ ],
175
+ "multi-gpu": [
176
+ "tests/models/cohere2/test_modeling_cohere2.py::Cohere2IntegrationTest::test_export_static_cache",
177
+ "tests/models/cohere2/test_modeling_cohere2.py::Cohere2IntegrationTest::test_model_fp16",
178
+ "tests/models/cohere2/test_modeling_cohere2.py::Cohere2IntegrationTest::test_model_pipeline_bf16"
179
+ ]
180
+ },
181
+ "colpali": {
182
+ "single-gpu": [
183
+ "tests/models/colpali/test_modeling_colpali.py::ColPaliModelIntegrationTest::test_model_integration_test"
184
+ ],
185
+ "multi-gpu": [
186
+ "tests/models/colpali/test_modeling_colpali.py::ColPaliModelIntegrationTest::test_model_integration_test"
187
+ ]
188
+ },
189
+ "ctrl": {
190
+ "single-gpu": [
191
+ "tests/models/ctrl/test_modeling_ctrl.py::CTRLModelLanguageGenerationTest::test_lm_generate_ctrl"
192
+ ],
193
+ "multi-gpu": [
194
+ "tests/models/ctrl/test_modeling_ctrl.py::CTRLModelLanguageGenerationTest::test_lm_generate_ctrl"
195
+ ]
196
+ },
197
+ "dab_detr": {
198
+ "single-gpu": [],
199
+ "multi-gpu": [
200
+ "tests/models/dab_detr/test_modeling_dab_detr.py::DabDetrModelTest::test_model_parallelism"
201
+ ]
202
+ },
203
+ "dac": {
204
+ "single-gpu": [
205
+ "tests/models/dac/test_modeling_dac.py::DacIntegrationTest::test_integration_16khz",
206
+ "tests/models/dac/test_modeling_dac.py::DacIntegrationTest::test_integration_24khz",
207
+ "tests/models/dac/test_modeling_dac.py::DacIntegrationTest::test_integration_44khz",
208
+ "tests/models/dac/test_modeling_dac.py::DacIntegrationTest::test_integration_batch_24khz",
209
+ "tests/models/dac/test_modeling_dac.py::DacIntegrationTest::test_integration_batch_44khz"
210
+ ],
211
+ "multi-gpu": [
212
+ "tests/models/dac/test_modeling_dac.py::DacIntegrationTest::test_integration_16khz",
213
+ "tests/models/dac/test_modeling_dac.py::DacIntegrationTest::test_integration_24khz",
214
+ "tests/models/dac/test_modeling_dac.py::DacIntegrationTest::test_integration_44khz",
215
+ "tests/models/dac/test_modeling_dac.py::DacIntegrationTest::test_integration_batch_24khz",
216
+ "tests/models/dac/test_modeling_dac.py::DacIntegrationTest::test_integration_batch_44khz"
217
+ ]
218
+ },
219
+ "data2vec": {
220
+ "single-gpu": [
221
+ "tests/models/data2vec/test_modeling_data2vec_vision.py::Data2VecVisionModelTest::test_sdpa_can_compile_dynamic"
222
+ ],
223
+ "multi-gpu": [
224
+ "tests/models/data2vec/test_modeling_data2vec_vision.py::Data2VecVisionModelTest::test_sdpa_can_compile_dynamic",
225
+ "tests/models/data2vec/test_modeling_data2vec_vision.py::Data2VecVisionModelTest::test_model_parallelism"
226
+ ]
227
+ },
228
+ "deepseek_v3": {
229
+ "single-gpu": [
230
+ "tests/models/deepseek_v3/test_modeling_deepseek_v3.py::DeepseekV3ModelTest::test_sdpa_can_dispatch_on_flash",
231
+ "tests/models/deepseek_v3/test_modeling_deepseek_v3.py::DeepseekV3ModelTest::test_torch_compile_for_training",
232
+ "tests/models/deepseek_v3/test_modeling_deepseek_v3.py::DeepseekV3IntegrationTest::test_compile_static_cache"
233
+ ],
234
+ "multi-gpu": [
235
+ "tests/models/deepseek_v3/test_modeling_deepseek_v3.py::DeepseekV3ModelTest::test_sdpa_can_dispatch_on_flash",
236
+ "tests/models/deepseek_v3/test_modeling_deepseek_v3.py::DeepseekV3ModelTest::test_torch_compile_for_training",
237
+ "tests/models/deepseek_v3/test_modeling_deepseek_v3.py::DeepseekV3IntegrationTest::test_compile_static_cache"
238
+ ]
239
+ },
240
+ "deformable_detr": {
241
+ "single-gpu": [
242
+ "tests/models/deformable_detr/test_modeling_deformable_detr.py::DeformableDetrModelIntegrationTests::test_inference_object_detection_head"
243
+ ],
244
+ "multi-gpu": [
245
+ "tests/models/deformable_detr/test_modeling_deformable_detr.py::DeformableDetrModelIntegrationTests::test_inference_object_detection_head"
246
+ ]
247
+ },
248
+ "depth_anything": {
249
+ "single-gpu": [
250
+ "tests/models/depth_anything/test_modeling_depth_anything.py::DepthAnythingModelIntegrationTest::test_inference"
251
+ ],
252
+ "multi-gpu": [
253
+ "tests/models/depth_anything/test_modeling_depth_anything.py::DepthAnythingModelIntegrationTest::test_inference",
254
+ "tests/models/depth_anything/test_modeling_depth_anything.py::DepthAnythingModelTest::test_model_parallelism"
255
+ ]
256
+ },
257
+ "depth_pro": {
258
+ "single-gpu": [
259
+ "tests/models/depth_pro/test_modeling_depth_pro.py::DepthProModelTest::test_sdpa_can_compile_dynamic"
260
+ ],
261
+ "multi-gpu": [
262
+ "tests/models/depth_pro/test_modeling_depth_pro.py::DepthProModelTest::test_sdpa_can_compile_dynamic"
263
+ ]
264
+ },
265
+ "diffllama": {
266
+ "single-gpu": [
267
+ "tests/models/diffllama/test_modeling_diffllama.py::DiffLlamaModelTest::test_sdpa_can_dispatch_on_flash",
268
+ "tests/models/diffllama/test_modeling_diffllama.py::DiffLlamaIntegrationTest::test_compile_static_cache"
269
+ ],
270
+ "multi-gpu": [
271
+ "tests/models/diffllama/test_modeling_diffllama.py::DiffLlamaModelTest::test_sdpa_can_dispatch_on_flash",
272
+ "tests/models/diffllama/test_modeling_diffllama.py::DiffLlamaIntegrationTest::test_compile_static_cache"
273
+ ]
274
+ },
275
+ "dpr": {
276
+ "single-gpu": [
277
+ "tests/models/dpr/test_modeling_dpr.py::DPRModelTest::test_sdpa_can_dispatch_on_flash"
278
+ ],
279
+ "multi-gpu": [
280
+ "tests/models/dpr/test_modeling_dpr.py::DPRModelTest::test_sdpa_can_dispatch_on_flash"
281
+ ]
282
+ },
283
+ "efficientnet": {
284
+ "single-gpu": [],
285
+ "multi-gpu": [
286
+ "tests/models/efficientnet/test_modeling_efficientnet.py::EfficientNetModelTest::test_model_parallelism"
287
+ ]
288
+ },
289
+ "emu3": {
290
+ "single-gpu": [
291
+ "tests/models/emu3/test_modeling_emu3.py::Emu3Vision2TextModelTest::test_attention_outputs",
292
+ "tests/models/emu3/test_modeling_emu3.py::Emu3Vision2TextModelTest::test_batching_equivalence",
293
+ "tests/models/emu3/test_modeling_emu3.py::Emu3Vision2TextModelTest::test_determinism",
294
+ "tests/models/emu3/test_modeling_emu3.py::Emu3Vision2TextModelTest::test_feed_forward_chunking",
295
+ "tests/models/emu3/test_modeling_emu3.py::Emu3Vision2TextModelTest::test_hidden_states_output",
296
+ "tests/models/emu3/test_modeling_emu3.py::Emu3Vision2TextModelTest::test_model_outputs_equivalence",
297
+ "tests/models/emu3/test_modeling_emu3.py::Emu3Vision2TextModelTest::test_resize_tokens_embeddings",
298
+ "tests/models/emu3/test_modeling_emu3.py::Emu3Vision2TextModelTest::test_retain_grad_hidden_states_attentions",
299
+ "tests/models/emu3/test_modeling_emu3.py::Emu3Vision2TextModelTest::test_save_load"
300
+ ],
301
+ "multi-gpu": [
302
+ "tests/models/emu3/test_modeling_emu3.py::Emu3Vision2TextModelTest::test_attention_outputs",
303
+ "tests/models/emu3/test_modeling_emu3.py::Emu3Vision2TextModelTest::test_batching_equivalence",
304
+ "tests/models/emu3/test_modeling_emu3.py::Emu3Vision2TextModelTest::test_determinism",
305
+ "tests/models/emu3/test_modeling_emu3.py::Emu3Vision2TextModelTest::test_feed_forward_chunking",
306
+ "tests/models/emu3/test_modeling_emu3.py::Emu3Vision2TextModelTest::test_hidden_states_output",
307
+ "tests/models/emu3/test_modeling_emu3.py::Emu3Vision2TextModelTest::test_model_outputs_equivalence",
308
+ "tests/models/emu3/test_modeling_emu3.py::Emu3Vision2TextModelTest::test_resize_tokens_embeddings",
309
+ "tests/models/emu3/test_modeling_emu3.py::Emu3Vision2TextModelTest::test_retain_grad_hidden_states_attentions",
310
+ "tests/models/emu3/test_modeling_emu3.py::Emu3Vision2TextModelTest::test_save_load",
311
+ "tests/models/emu3/test_modeling_emu3.py::Emu3Vision2TextModelTest::test_model_parallelism",
312
+ "tests/models/emu3/test_modeling_emu3.py::Emu3Vision2TextModelTest::test_multi_gpu_data_parallel_forward"
313
+ ]
314
+ },
315
+ "encodec": {
316
+ "single-gpu": [
317
+ "tests/models/encodec/test_modeling_encodec.py::EncodecIntegrationTest::test_batch_48kHz"
318
+ ],
319
+ "multi-gpu": [
320
+ "tests/models/encodec/test_modeling_encodec.py::EncodecIntegrationTest::test_batch_48kHz"
321
+ ]
322
+ },
323
+ "encoder_decoder": {
324
+ "single-gpu": [
325
+ "tests/models/encoder_decoder/test_modeling_encoder_decoder.py::BertGenerationEncoderDecoderModelTest::test_roberta2roberta_summarization",
326
+ "tests/models/encoder_decoder/test_modeling_encoder_decoder.py::GPT2EncoderDecoderModelTest::test_bert2gpt2_summarization"
327
+ ],
328
+ "multi-gpu": [
329
+ "tests/models/encoder_decoder/test_modeling_encoder_decoder.py::BertGenerationEncoderDecoderModelTest::test_roberta2roberta_summarization",
330
+ "tests/models/encoder_decoder/test_modeling_encoder_decoder.py::GPT2EncoderDecoderModelTest::test_bert2gpt2_summarization"
331
+ ]
332
+ },
333
+ "falcon": {
334
+ "single-gpu": [
335
+ "tests/models/falcon/test_modeling_falcon.py::FalconLanguageGenerationTest::test_falcon_alibi_sdpa_matches_eager"
336
+ ],
337
+ "multi-gpu": [
338
+ "tests/models/falcon/test_modeling_falcon.py::FalconLanguageGenerationTest::test_falcon_alibi_sdpa_matches_eager"
339
+ ]
340
+ },
341
+ "falcon_h1": {
342
+ "single-gpu": [
343
+ "tests/models/falcon_h1/test_modeling_falcon_h1.py::FalconH1ModelIntegrationTest::test_falcon_h1_hard"
344
+ ],
345
+ "multi-gpu": [
346
+ "tests/models/falcon_h1/test_modeling_falcon_h1.py::FalconH1ModelIntegrationTest::test_falcon_h1_hard"
347
+ ]
348
+ },
349
+ "falcon_mamba": {
350
+ "single-gpu": [],
351
+ "multi-gpu": [
352
+ "tests/models/falcon_mamba/test_modeling_falcon_mamba.py::FalconMambaModelTest::test_multi_gpu_data_parallel_forward"
353
+ ]
354
+ },
355
+ "fastspeech2_conformer": {
356
+ "single-gpu": [
357
+ "tests/models/fastspeech2_conformer/test_modeling_fastspeech2_conformer.py::FastSpeech2ConformerWithHifiGanTest::test_model_outputs_equivalence"
358
+ ],
359
+ "multi-gpu": [
360
+ "tests/models/fastspeech2_conformer/test_modeling_fastspeech2_conformer.py::FastSpeech2ConformerModelTest::test_multi_gpu_data_parallel_forward"
361
+ ]
362
+ },
363
+ "flava": {
364
+ "single-gpu": [
365
+ "tests/models/flava/test_modeling_flava.py::FlavaModelIntegrationTest::test_inference",
366
+ "tests/models/flava/test_modeling_flava.py::FlavaForPreTrainingIntegrationTest::test_inference_with_itm_labels"
367
+ ],
368
+ "multi-gpu": [
369
+ "tests/models/flava/test_modeling_flava.py::FlavaModelIntegrationTest::test_inference",
370
+ "tests/models/flava/test_modeling_flava.py::FlavaForPreTrainingIntegrationTest::test_inference_with_itm_labels"
371
+ ]
372
+ },
373
+ "gemma": {
374
+ "single-gpu": [
375
+ "tests/models/gemma/test_modeling_gemma.py::GemmaModelTest::test_torch_compile_for_training",
376
+ "tests/models/gemma/test_modeling_gemma.py::GemmaIntegrationTest::test_model_2b_bf16_dola",
377
+ "tests/models/gemma/test_modeling_gemma.py::GemmaIntegrationTest::test_model_7b_bf16",
378
+ "tests/models/gemma/test_modeling_gemma.py::GemmaIntegrationTest::test_model_7b_fp16",
379
+ "tests/models/gemma/test_modeling_gemma.py::GemmaIntegrationTest::test_model_7b_fp16_static_cache"
380
+ ],
381
+ "multi-gpu": [
382
+ "tests/models/gemma/test_modeling_gemma.py::GemmaModelTest::test_torch_compile_for_training",
383
+ "tests/models/gemma/test_modeling_gemma.py::GemmaIntegrationTest::test_model_2b_bf16_dola",
384
+ "tests/models/gemma/test_modeling_gemma.py::GemmaIntegrationTest::test_model_7b_bf16",
385
+ "tests/models/gemma/test_modeling_gemma.py::GemmaIntegrationTest::test_model_7b_fp16"
386
+ ]
387
+ },
388
+ "gemma2": {
389
+ "single-gpu": [
390
+ "tests/models/gemma2/test_modeling_gemma2.py::Gemma2IntegrationTest::test_export_static_cache"
391
+ ],
392
+ "multi-gpu": [
393
+ "tests/models/gemma2/test_modeling_gemma2.py::Gemma2IntegrationTest::test_export_static_cache"
394
+ ]
395
+ },
396
+ "gemma3": {
397
+ "single-gpu": [
398
+ "tests/models/gemma3/test_modeling_gemma3.py::Gemma3IntegrationTest::test_model_4b_batch",
399
+ "tests/models/gemma3/test_modeling_gemma3.py::Gemma3IntegrationTest::test_model_4b_batch_crops",
400
+ "tests/models/gemma3/test_modeling_gemma3.py::Gemma3IntegrationTest::test_model_4b_bf16",
401
+ "tests/models/gemma3/test_modeling_gemma3.py::Gemma3IntegrationTest::test_model_4b_multiimage"
402
+ ],
403
+ "multi-gpu": [
404
+ "tests/models/gemma3/test_modeling_gemma3.py::Gemma3IntegrationTest::test_model_4b_batch",
405
+ "tests/models/gemma3/test_modeling_gemma3.py::Gemma3IntegrationTest::test_model_4b_batch_crops",
406
+ "tests/models/gemma3/test_modeling_gemma3.py::Gemma3IntegrationTest::test_model_4b_bf16"
407
+ ]
408
+ },
409
+ "glm": {
410
+ "single-gpu": [
411
+ "tests/models/glm/test_modeling_glm.py::GlmIntegrationTest::test_model_9b_eager"
412
+ ],
413
+ "multi-gpu": [
414
+ "tests/models/glm/test_modeling_glm.py::GlmIntegrationTest::test_model_9b_eager"
415
+ ]
416
+ },
417
+ "glm4": {
418
+ "single-gpu": [
419
+ "tests/models/glm4/test_modeling_glm4.py::Glm4IntegrationTest::test_model_9b_eager"
420
+ ],
421
+ "multi-gpu": [
422
+ "tests/models/glm4/test_modeling_glm4.py::Glm4IntegrationTest::test_model_9b_eager"
423
+ ]
424
+ },
425
+ "glpn": {
426
+ "single-gpu": [],
427
+ "multi-gpu": [
428
+ "tests/models/glpn/test_modeling_glpn.py::GLPNModelTest::test_model_parallelism"
429
+ ]
430
+ },
431
+ "got_ocr2": {
432
+ "single-gpu": [
433
+ "tests/models/got_ocr2/test_modeling_got_ocr2.py::GotOcr2ModelTest::test_generate_continue_from_past_key_values"
434
+ ],
435
+ "multi-gpu": [
436
+ "tests/models/got_ocr2/test_modeling_got_ocr2.py::GotOcr2ModelTest::test_generate_continue_from_past_key_values"
437
+ ]
438
+ },
439
+ "gpt_bigcode": {
440
+ "single-gpu": [
441
+ "tests/models/gpt_bigcode/test_modeling_gpt_bigcode.py::GPTBigCodeModelTest::test_sdpa_can_dispatch_on_flash",
442
+ "tests/models/gpt_bigcode/test_modeling_gpt_bigcode.py::GPTBigCodeMHAModelTest::test_sdpa_can_dispatch_on_flash",
443
+ "tests/models/gpt_bigcode/test_modeling_gpt_bigcode.py::GPTBigCodeModelLanguageGenerationTest::test_generate_batched",
444
+ "tests/models/gpt_bigcode/test_modeling_gpt_bigcode.py::GPTBigCodeModelLanguageGenerationTest::test_generate_simple"
445
+ ],
446
+ "multi-gpu": [
447
+ "tests/models/gpt_bigcode/test_modeling_gpt_bigcode.py::GPTBigCodeModelTest::test_sdpa_can_dispatch_on_flash",
448
+ "tests/models/gpt_bigcode/test_modeling_gpt_bigcode.py::GPTBigCodeMHAModelTest::test_sdpa_can_dispatch_on_flash",
449
+ "tests/models/gpt_bigcode/test_modeling_gpt_bigcode.py::GPTBigCodeModelLanguageGenerationTest::test_generate_batched",
450
+ "tests/models/gpt_bigcode/test_modeling_gpt_bigcode.py::GPTBigCodeModelLanguageGenerationTest::test_generate_simple",
451
+ "tests/models/gpt_bigcode/test_modeling_gpt_bigcode.py::GPTBigCodeModelTest::test_model_parallelism",
452
+ "tests/models/gpt_bigcode/test_modeling_gpt_bigcode.py::GPTBigCodeMHAModelTest::test_model_parallelism"
453
+ ]
454
+ },
455
+ "gpt_neo": {
456
+ "single-gpu": [
457
+ "tests/models/gpt_neo/test_modeling_gpt_neo.py::GPTNeoModelLanguageGenerationTest::test_batch_generation",
458
+ "tests/models/gpt_neo/test_modeling_gpt_neo.py::GPTNeoModelLanguageGenerationTest::test_gpt_neo_sample",
459
+ "tests/models/gpt_neo/test_modeling_gpt_neo.py::GPTNeoModelLanguageGenerationTest::test_lm_generate_gpt_neo"
460
+ ],
461
+ "multi-gpu": [
462
+ "tests/models/gpt_neo/test_modeling_gpt_neo.py::GPTNeoModelLanguageGenerationTest::test_batch_generation",
463
+ "tests/models/gpt_neo/test_modeling_gpt_neo.py::GPTNeoModelLanguageGenerationTest::test_gpt_neo_sample",
464
+ "tests/models/gpt_neo/test_modeling_gpt_neo.py::GPTNeoModelLanguageGenerationTest::test_lm_generate_gpt_neo"
465
+ ]
466
+ },
467
+ "gpt_neox": {
468
+ "single-gpu": [
469
+ "tests/models/gpt_neox/test_modeling_gpt_neox.py::GPTNeoXLanguageGenerationTest::test_lm_generate_flex_attn_gptneox"
470
+ ],
471
+ "multi-gpu": [
472
+ "tests/models/gpt_neox/test_modeling_gpt_neox.py::GPTNeoXLanguageGenerationTest::test_lm_generate_flex_attn_gptneox"
473
+ ]
474
+ },
475
+ "granite": {
476
+ "single-gpu": [
477
+ "tests/models/granite/test_modeling_granite.py::GraniteIntegrationTest::test_model_3b_logits_bf16"
478
+ ],
479
+ "multi-gpu": [
480
+ "tests/models/granite/test_modeling_granite.py::GraniteIntegrationTest::test_model_3b_logits_bf16"
481
+ ]
482
+ },
483
+ "grounding_dino": {
484
+ "single-gpu": [
485
+ "tests/models/grounding_dino/test_modeling_grounding_dino.py::GroundingDinoModelIntegrationTests::test_cross_attention_mask",
486
+ "tests/models/grounding_dino/test_modeling_grounding_dino.py::GroundingDinoModelIntegrationTests::test_grounding_dino_loss",
487
+ "tests/models/grounding_dino/test_modeling_grounding_dino.py::GroundingDinoModelIntegrationTests::test_inference_object_detection_head_equivalence_cpu_accelerator"
488
+ ],
489
+ "multi-gpu": [
490
+ "tests/models/grounding_dino/test_modeling_grounding_dino.py::GroundingDinoModelIntegrationTests::test_cross_attention_mask",
491
+ "tests/models/grounding_dino/test_modeling_grounding_dino.py::GroundingDinoModelIntegrationTests::test_grounding_dino_loss"
492
+ ]
493
+ },
494
+ "helium": {
495
+ "single-gpu": [
496
+ "tests/models/helium/test_modeling_helium.py::GemmaModelTest::test_torch_compile_for_training",
497
+ "tests/models/helium/test_modeling_helium.py::HeliumModelTest::test_torch_compile_for_training",
498
+ "tests/models/helium/test_modeling_helium.py::HeliumIntegrationTest::test_model_2b"
499
+ ],
500
+ "multi-gpu": [
501
+ "tests/models/helium/test_modeling_helium.py::GemmaModelTest::test_torch_compile_for_training",
502
+ "tests/models/helium/test_modeling_helium.py::HeliumModelTest::test_torch_compile_for_training",
503
+ "tests/models/helium/test_modeling_helium.py::HeliumIntegrationTest::test_model_2b"
504
+ ]
505
+ },
506
+ "hiera": {
507
+ "single-gpu": [
508
+ "tests/models/hiera/test_modeling_hiera.py::HieraModelIntegrationTest::test_inference_image_classification_head"
509
+ ],
510
+ "multi-gpu": [
511
+ "tests/models/hiera/test_modeling_hiera.py::HieraModelIntegrationTest::test_inference_image_classification_head"
512
+ ]
513
+ },
514
+ "hubert": {
515
+ "single-gpu": [
516
+ "tests/models/hubert/test_modeling_hubert.py::HubertModelIntegrationTest::test_inference_hubert_25hz"
517
+ ],
518
+ "multi-gpu": [
519
+ "tests/models/hubert/test_modeling_hubert.py::HubertModelIntegrationTest::test_inference_hubert_25hz"
520
+ ]
521
+ },
522
+ "instructblip": {
523
+ "single-gpu": [
524
+ "tests/models/instructblip/test_modeling_instructblip.py::InstructBlipForConditionalGenerationDecoderOnlyTest::test_eager_matches_sdpa_generate",
525
+ "tests/models/instructblip/test_modeling_instructblip.py::InstructBlipModelIntegrationTest::test_inference_flant5_xl"
526
+ ],
527
+ "multi-gpu": [
528
+ "tests/models/instructblip/test_modeling_instructblip.py::InstructBlipForConditionalGenerationDecoderOnlyTest::test_eager_matches_sdpa_generate",
529
+ "tests/models/instructblip/test_modeling_instructblip.py::InstructBlipModelIntegrationTest::test_inference_flant5_xl",
530
+ "tests/models/instructblip/test_modeling_instructblip.py::InstructBlipForConditionalGenerationDecoderOnlyTest::test_model_parallelism"
531
+ ]
532
+ },
533
+ "instructblipvideo": {
534
+ "single-gpu": [
535
+ "tests/models/instructblipvideo/test_modeling_instructblipvideo.py::InstructBlipVideoForConditionalGenerationDecoderOnlyTest::test_eager_matches_sdpa_generate"
536
+ ],
537
+ "multi-gpu": [
538
+ "tests/models/instructblipvideo/test_modeling_instructblipvideo.py::InstructBlipVideoForConditionalGenerationDecoderOnlyTest::test_eager_matches_sdpa_generate",
539
+ "tests/models/instructblipvideo/test_modeling_instructblipvideo.py::InstructBlipVideoForConditionalGenerationDecoderOnlyTest::test_model_parallelism"
540
+ ]
541
+ },
542
+ "internvl": {
543
+ "single-gpu": [
544
+ "tests/models/internvl/test_modeling_internvl.py::InternVLLlamaIntegrationTest::test_llama_small_model_integration_forward",
545
+ "tests/models/internvl/test_modeling_internvl.py::InternVLLlamaIntegrationTest::test_llama_small_model_integration_interleaved_images_videos",
546
+ "tests/models/internvl/test_video_processor_internvl.py::InternVLVideoProcessingTest::test_can_compile_fast_video_processor"
547
+ ],
548
+ "multi-gpu": [
549
+ "tests/models/internvl/test_modeling_internvl.py::InternVLLlamaIntegrationTest::test_llama_small_model_integration_forward",
550
+ "tests/models/internvl/test_modeling_internvl.py::InternVLLlamaIntegrationTest::test_llama_small_model_integration_interleaved_images_videos",
551
+ "tests/models/internvl/test_video_processor_internvl.py::InternVLVideoProcessingTest::test_can_compile_fast_video_processor"
552
+ ]
553
+ },
554
+ "jamba": {
555
+ "single-gpu": [
556
+ "tests/models/jamba/test_modeling_jamba.py::JambaModelTest::test_sdpa_can_dispatch_on_flash",
557
+ "tests/models/jamba/test_modeling_jamba.py::JambaModelIntegrationTest::test_simple_batched_generate_with_padding",
558
+ "tests/models/jamba/test_modeling_jamba.py::JambaModelIntegrationTest::test_simple_generate"
559
+ ],
560
+ "multi-gpu": [
561
+ "tests/models/jamba/test_modeling_jamba.py::JambaModelTest::test_sdpa_can_dispatch_on_flash",
562
+ "tests/models/jamba/test_modeling_jamba.py::JambaModelIntegrationTest::test_simple_batched_generate_with_padding",
563
+ "tests/models/jamba/test_modeling_jamba.py::JambaModelIntegrationTest::test_simple_generate"
564
+ ]
565
+ },
566
+ "jetmoe": {
567
+ "single-gpu": [
568
+ "tests/models/jetmoe/test_modeling_jetmoe.py::JetMoeIntegrationTest::test_model_8b_batched_generation",
569
+ "tests/models/jetmoe/test_modeling_jetmoe.py::JetMoeIntegrationTest::test_model_8b_generation",
570
+ "tests/models/jetmoe/test_modeling_jetmoe.py::JetMoeIntegrationTest::test_model_8b_logits"
571
+ ],
572
+ "multi-gpu": [
573
+ "tests/models/jetmoe/test_modeling_jetmoe.py::JetMoeIntegrationTest::test_model_8b_batched_generation",
574
+ "tests/models/jetmoe/test_modeling_jetmoe.py::JetMoeIntegrationTest::test_model_8b_generation",
575
+ "tests/models/jetmoe/test_modeling_jetmoe.py::JetMoeIntegrationTest::test_model_8b_logits"
576
+ ]
577
+ },
578
+ "kosmos2": {
579
+ "single-gpu": [
580
+ "tests/models/kosmos2/test_modeling_kosmos2.py::Kosmos2ModelTest::test_sdpa_can_dispatch_on_flash",
581
+ "tests/models/kosmos2/test_modeling_kosmos2.py::Kosmos2ModelIntegrationTest::test_inference_interpolate_pos_encoding",
582
+ "tests/models/kosmos2/test_modeling_kosmos2.py::Kosmos2ModelIntegrationTest::test_snowman_image_captioning",
583
+ "tests/models/kosmos2/test_modeling_kosmos2.py::Kosmos2ModelIntegrationTest::test_snowman_image_captioning_batch"
584
+ ],
585
+ "multi-gpu": [
586
+ "tests/models/kosmos2/test_modeling_kosmos2.py::Kosmos2ModelTest::test_sdpa_can_dispatch_on_flash",
587
+ "tests/models/kosmos2/test_modeling_kosmos2.py::Kosmos2ModelIntegrationTest::test_inference_interpolate_pos_encoding",
588
+ "tests/models/kosmos2/test_modeling_kosmos2.py::Kosmos2ModelIntegrationTest::test_snowman_image_captioning",
589
+ "tests/models/kosmos2/test_modeling_kosmos2.py::Kosmos2ModelIntegrationTest::test_snowman_image_captioning_batch"
590
+ ]
591
+ },
592
+ "led": {
593
+ "single-gpu": [
594
+ "tests/models/led/test_modeling_led.py::LEDModelIntegrationTests::test_inference_head",
595
+ "tests/models/led/test_modeling_led.py::LEDModelIntegrationTests::test_inference_no_head"
596
+ ],
597
+ "multi-gpu": [
598
+ "tests/models/led/test_modeling_led.py::LEDModelIntegrationTests::test_inference_head",
599
+ "tests/models/led/test_modeling_led.py::LEDModelIntegrationTests::test_inference_no_head"
600
+ ]
601
+ },
602
+ "lilt": {
603
+ "single-gpu": [
604
+ "tests/models/lilt/test_modeling_lilt.py::LiltModelTest::test_model_outputs_equivalence"
605
+ ],
606
+ "multi-gpu": [
607
+ "tests/models/lilt/test_modeling_lilt.py::LiltModelTest::test_model_outputs_equivalence",
608
+ "tests/models/lilt/test_modeling_lilt.py::LiltModelTest::test_model_parallelism"
609
+ ]
610
+ },
611
+ "llama": {
612
+ "single-gpu": [
613
+ "tests/models/llama/test_modeling_llama.py::LlamaModelTest::test_torch_compile_for_training",
614
+ "tests/models/llama/test_modeling_llama.py::LlamaIntegrationTest::test_model_7b_dola_generation",
615
+ "tests/models/llama/test_modeling_llama.py::LlamaIntegrationTest::test_model_7b_logits_bf16"
616
+ ],
617
+ "multi-gpu": [
618
+ "tests/models/llama/test_modeling_llama.py::LlamaModelTest::test_torch_compile_for_training",
619
+ "tests/models/llama/test_modeling_llama.py::LlamaIntegrationTest::test_model_7b_dola_generation",
620
+ "tests/models/llama/test_modeling_llama.py::LlamaIntegrationTest::test_model_7b_logits_bf16"
621
+ ]
622
+ },
623
+ "llava": {
624
+ "single-gpu": [
625
+ "tests/models/llava/test_modeling_llava.py::LlavaForConditionalGenerationIntegrationTest::test_batched_generation"
626
+ ],
627
+ "multi-gpu": [
628
+ "tests/models/llava/test_modeling_llava.py::LlavaForConditionalGenerationIntegrationTest::test_batched_generation"
629
+ ]
630
+ },
631
+ "longt5": {
632
+ "single-gpu": [
633
+ "tests/models/longt5/test_modeling_longt5.py::LongT5ModelIntegrationTests::test_inference_hidden_states",
634
+ "tests/models/longt5/test_modeling_longt5.py::LongT5ModelIntegrationTests::test_summarization"
635
+ ],
636
+ "multi-gpu": [
637
+ "tests/models/longt5/test_modeling_longt5.py::LongT5ModelIntegrationTests::test_inference_hidden_states",
638
+ "tests/models/longt5/test_modeling_longt5.py::LongT5ModelIntegrationTests::test_summarization"
639
+ ]
640
+ },
641
+ "mamba": {
642
+ "single-gpu": [
643
+ "tests/models/mamba/test_modeling_mamba.py::MambaIntegrationTests::test_compile_mamba_cache"
644
+ ],
645
+ "multi-gpu": [
646
+ "tests/models/mamba/test_modeling_mamba.py::MambaIntegrationTests::test_compile_mamba_cache",
647
+ "tests/models/mamba/test_modeling_mamba.py::MambaModelTest::test_multi_gpu_data_parallel_forward"
648
+ ]
649
+ },
650
+ "mamba2": {
651
+ "single-gpu": [
652
+ "tests/models/mamba2/test_modeling_mamba2.py::Mamba2ModelTest::test_mamba2_caching",
653
+ "tests/models/mamba2/test_modeling_mamba2.py::Mamba2IntegrationTest::test_simple_generate"
654
+ ],
655
+ "multi-gpu": [
656
+ "tests/models/mamba2/test_modeling_mamba2.py::Mamba2ModelTest::test_mamba2_caching",
657
+ "tests/models/mamba2/test_modeling_mamba2.py::Mamba2IntegrationTest::test_simple_generate"
658
+ ]
659
+ },
660
+ "mgp_str": {
661
+ "single-gpu": [],
662
+ "multi-gpu": [
663
+ "tests/models/mgp_str/test_modeling_mgp_str.py::MgpstrModelTest::test_model_parallelism"
664
+ ]
665
+ },
666
+ "mimi": {
667
+ "single-gpu": [],
668
+ "multi-gpu": [
669
+ "tests/models/mimi/test_modeling_mimi.py::MimiModelTest::test_model_parallelism"
670
+ ]
671
+ },
672
+ "minimax": {
673
+ "single-gpu": [
674
+ "tests/models/minimax/test_modeling_minimax.py::MiniMaxIntegrationTest::test_small_model_logits"
675
+ ],
676
+ "multi-gpu": [
677
+ "tests/models/minimax/test_modeling_minimax.py::MiniMaxIntegrationTest::test_small_model_logits",
678
+ "tests/models/minimax/test_modeling_minimax.py::MiniMaxModelTest::test_multi_gpu_data_parallel_forward"
679
+ ]
680
+ },
681
+ "mistral": {
682
+ "single-gpu": [
683
+ "tests/models/mistral/test_modeling_mistral.py::MistralIntegrationTest::test_model_7b_logits"
684
+ ],
685
+ "multi-gpu": [
686
+ "tests/models/mistral/test_modeling_mistral.py::MistralIntegrationTest::test_model_7b_logits"
687
+ ]
688
+ },
689
+ "mistral3": {
690
+ "single-gpu": [
691
+ "tests/models/mistral3/test_modeling_mistral3.py::Mistral3IntegrationTest::test_mistral3_integration_batched_generate",
692
+ "tests/models/mistral3/test_modeling_mistral3.py::Mistral3IntegrationTest::test_mistral3_integration_batched_generate_multi_image",
693
+ "tests/models/mistral3/test_modeling_mistral3.py::Mistral3IntegrationTest::test_mistral3_integration_generate"
694
+ ],
695
+ "multi-gpu": [
696
+ "tests/models/mistral3/test_modeling_mistral3.py::Mistral3IntegrationTest::test_mistral3_integration_batched_generate",
697
+ "tests/models/mistral3/test_modeling_mistral3.py::Mistral3IntegrationTest::test_mistral3_integration_batched_generate_multi_image",
698
+ "tests/models/mistral3/test_modeling_mistral3.py::Mistral3IntegrationTest::test_mistral3_integration_generate"
699
+ ]
700
+ },
701
+ "mixtral": {
702
+ "single-gpu": [
703
+ "tests/models/mixtral/test_modeling_mixtral.py::MixtralIntegrationTest::test_small_model_logits",
704
+ "tests/models/mixtral/test_modeling_mixtral.py::MixtralIntegrationTest::test_small_model_logits_batched"
705
+ ],
706
+ "multi-gpu": [
707
+ "tests/models/mixtral/test_modeling_mixtral.py::MixtralIntegrationTest::test_small_model_logits",
708
+ "tests/models/mixtral/test_modeling_mixtral.py::MixtralIntegrationTest::test_small_model_logits_batched"
709
+ ]
710
+ },
711
+ "mlcd": {
712
+ "single-gpu": [
713
+ "tests/models/mlcd/test_modeling_mlcd.py::MLCDVisionModelIntegrationTest::test_inference"
714
+ ],
715
+ "multi-gpu": [
716
+ "tests/models/mlcd/test_modeling_mlcd.py::MLCDVisionModelIntegrationTest::test_inference"
717
+ ]
718
+ },
719
+ "mllama": {
720
+ "single-gpu": [
721
+ "tests/models/mllama/test_modeling_mllama.py::MllamaForConditionalGenerationModelTest::test_sdpa_can_dispatch_on_flash"
722
+ ],
723
+ "multi-gpu": [
724
+ "tests/models/mllama/test_modeling_mllama.py::MllamaForConditionalGenerationModelTest::test_sdpa_can_dispatch_on_flash"
725
+ ]
726
+ },
727
+ "mobilenet_v2": {
728
+ "single-gpu": [],
729
+ "multi-gpu": [
730
+ "tests/models/mobilenet_v2/test_modeling_mobilenet_v2.py::MobileNetV2ModelTest::test_model_parallelism"
731
+ ]
732
+ },
733
+ "mobilevit": {
734
+ "single-gpu": [],
735
+ "multi-gpu": [
736
+ "tests/models/mobilevit/test_modeling_mobilevit.py::MobileViTModelTest::test_model_parallelism"
737
+ ]
738
+ },
739
+ "mobilevitv2": {
740
+ "single-gpu": [],
741
+ "multi-gpu": [
742
+ "tests/models/mobilevitv2/test_modeling_mobilevitv2.py::MobileViTV2ModelTest::test_model_parallelism"
743
+ ]
744
+ },
745
+ "moonshine": {
746
+ "single-gpu": [
747
+ "tests/models/moonshine/test_modeling_moonshine.py::MoonshineModelIntegrationTests::test_base_logits_batch",
748
+ "tests/models/moonshine/test_modeling_moonshine.py::MoonshineModelIntegrationTests::test_tiny_logits_batch"
749
+ ],
750
+ "multi-gpu": [
751
+ "tests/models/moonshine/test_modeling_moonshine.py::MoonshineModelIntegrationTests::test_base_logits_batch",
752
+ "tests/models/moonshine/test_modeling_moonshine.py::MoonshineModelIntegrationTests::test_tiny_logits_batch",
753
+ "tests/models/moonshine/test_modeling_moonshine.py::MoonshineModelTest::test_multi_gpu_data_parallel_forward",
754
+ "tests/models/moonshine/test_modeling_moonshine.py::MoonshineModelTest::test_sdpa_can_compile_dynamic"
755
+ ]
756
+ },
757
+ "moshi": {
758
+ "single-gpu": [
759
+ "tests/models/moshi/test_modeling_moshi.py::MoshiIntegrationTests::test_moshika_conditional_greedy",
760
+ "tests/models/moshi/test_modeling_moshi.py::MoshiIntegrationTests::test_moshika_greedy_unconditional_fp16",
761
+ "tests/models/moshi/test_modeling_moshi.py::MoshiIntegrationTests::test_moshiko_greedy_unconditional_fp16",
762
+ "tests/models/moshi/test_modeling_moshi.py::MoshiIntegrationTests::test_moshiko_greedy_unconditional_fp16_eager",
763
+ "tests/models/moshi/test_modeling_moshi.py::MoshiIntegrationTests::test_moshiko_greedy_unconditional_fp32"
764
+ ],
765
+ "multi-gpu": [
766
+ "tests/models/moshi/test_modeling_moshi.py::MoshiIntegrationTests::test_moshika_conditional_greedy",
767
+ "tests/models/moshi/test_modeling_moshi.py::MoshiIntegrationTests::test_moshika_greedy_unconditional_fp16",
768
+ "tests/models/moshi/test_modeling_moshi.py::MoshiIntegrationTests::test_moshiko_greedy_unconditional_fp16",
769
+ "tests/models/moshi/test_modeling_moshi.py::MoshiIntegrationTests::test_moshiko_greedy_unconditional_fp16_eager",
770
+ "tests/models/moshi/test_modeling_moshi.py::MoshiIntegrationTests::test_moshiko_greedy_unconditional_fp32"
771
+ ]
772
+ },
773
+ "mvp": {
774
+ "single-gpu": [
775
+ "tests/models/mvp/test_modeling_mvp.py::MvpModelIntegrationTests::test_inference_no_head"
776
+ ],
777
+ "multi-gpu": [
778
+ "tests/models/mvp/test_modeling_mvp.py::MvpModelIntegrationTests::test_inference_no_head"
779
+ ]
780
+ },
781
+ "nemotron": {
782
+ "single-gpu": [
783
+ "tests/models/nemotron/test_modeling_nemotron.py::NemotronModelTest::test_torch_compile_for_training",
784
+ "tests/models/nemotron/test_modeling_nemotron.py::NemotronIntegrationTest::test_nemotron_8b_generation_eager",
785
+ "tests/models/nemotron/test_modeling_nemotron.py::NemotronIntegrationTest::test_nemotron_8b_generation_fa2"
786
+ ],
787
+ "multi-gpu": [
788
+ "tests/models/nemotron/test_modeling_nemotron.py::NemotronModelTest::test_torch_compile_for_training",
789
+ "tests/models/nemotron/test_modeling_nemotron.py::NemotronIntegrationTest::test_nemotron_8b_generation_eager",
790
+ "tests/models/nemotron/test_modeling_nemotron.py::NemotronIntegrationTest::test_nemotron_8b_generation_fa2"
791
+ ]
792
+ },
793
+ "olmo": {
794
+ "single-gpu": [
795
+ "tests/models/olmo/test_modeling_olmo.py::OlmoIntegrationTest::test_model_1b_logits",
796
+ "tests/models/olmo/test_modeling_olmo.py::OlmoIntegrationTest::test_model_7b_greedy_generation",
797
+ "tests/models/olmo/test_modeling_olmo.py::OlmoIntegrationTest::test_model_7b_logits",
798
+ "tests/models/olmo/test_modeling_olmo.py::OlmoIntegrationTest::test_model_7b_twin_2t_logits"
799
+ ],
800
+ "multi-gpu": []
801
+ },
802
+ "olmo2": {
803
+ "single-gpu": [
804
+ "tests/models/olmo2/test_modeling_olmo2.py::Olmo2IntegrationTest::test_model_1b_logits_bfloat16",
805
+ "tests/models/olmo2/test_modeling_olmo2.py::Olmo2IntegrationTest::test_model_7b_logits"
806
+ ],
807
+ "multi-gpu": [
808
+ "tests/models/olmo2/test_modeling_olmo2.py::Olmo2IntegrationTest::test_model_1b_logits_bfloat16"
809
+ ]
810
+ },
811
+ "olmoe": {
812
+ "single-gpu": [
813
+ "tests/models/olmoe/test_modeling_olmoe.py::OlmoeIntegrationTest::test_model_7b_greedy_generation",
814
+ "tests/models/olmoe/test_modeling_olmoe.py::OlmoeIntegrationTest::test_model_7b_logits"
815
+ ],
816
+ "multi-gpu": [
817
+ "tests/models/olmoe/test_modeling_olmoe.py::OlmoeIntegrationTest::test_model_7b_logits"
818
+ ]
819
+ },
820
+ "omdet_turbo": {
821
+ "single-gpu": [
822
+ "tests/models/omdet_turbo/test_modeling_omdet_turbo.py::OmDetTurboModelIntegrationTests::test_inference_object_detection_head_fp16"
823
+ ],
824
+ "multi-gpu": []
825
+ },
826
+ "openai": {
827
+ "single-gpu": [
828
+ "tests/models/openai/test_modeling_openai.py::OPENAIGPTModelLanguageGenerationTest::test_lm_generate_openai_gpt"
829
+ ],
830
+ "multi-gpu": [
831
+ "tests/models/openai/test_modeling_openai.py::OPENAIGPTModelLanguageGenerationTest::test_lm_generate_openai_gpt"
832
+ ]
833
+ },
834
+ "opt": {
835
+ "single-gpu": [
836
+ "tests/models/opt/test_modeling_opt.py::OPTGenerationTest::test_batch_generation"
837
+ ],
838
+ "multi-gpu": [
839
+ "tests/models/opt/test_modeling_opt.py::OPTGenerationTest::test_batch_generation"
840
+ ]
841
+ },
842
+ "paligemma": {
843
+ "single-gpu": [
844
+ "tests/models/paligemma/test_modeling_paligemma.py::PaliGemmaForConditionalGenerationIntegrationTest::test_integration_detection_bug"
845
+ ],
846
+ "multi-gpu": [
847
+ "tests/models/paligemma/test_modeling_paligemma.py::PaliGemmaForConditionalGenerationIntegrationTest::test_integration_detection_bug"
848
+ ]
849
+ },
850
+ "patchtst": {
851
+ "single-gpu": [
852
+ "tests/models/patchtst/test_modeling_patchtst.py::PatchTSTModelIntegrationTests::test_prediction_generation",
853
+ "tests/models/patchtst/test_modeling_patchtst.py::PatchTSTModelIntegrationTests::test_prediction_head"
854
+ ],
855
+ "multi-gpu": [
856
+ "tests/models/patchtst/test_modeling_patchtst.py::PatchTSTModelIntegrationTests::test_prediction_generation",
857
+ "tests/models/patchtst/test_modeling_patchtst.py::PatchTSTModelIntegrationTests::test_prediction_head"
858
+ ]
859
+ },
860
+ "perceiver": {
861
+ "single-gpu": [
862
+ "tests/models/perceiver/test_modeling_perceiver.py::PerceiverModelIntegrationTest::test_inference_interpolate_pos_encoding",
863
+ "tests/models/perceiver/test_modeling_perceiver.py::PerceiverModelIntegrationTest::test_inference_optical_flow"
864
+ ],
865
+ "multi-gpu": [
866
+ "tests/models/perceiver/test_modeling_perceiver.py::PerceiverModelIntegrationTest::test_inference_interpolate_pos_encoding",
867
+ "tests/models/perceiver/test_modeling_perceiver.py::PerceiverModelIntegrationTest::test_inference_optical_flow"
868
+ ]
869
+ },
870
+ "phi3": {
871
+ "single-gpu": [
872
+ "tests/models/phi3/test_modeling_phi3.py::Phi3IntegrationTest::test_export_static_cache",
873
+ "tests/models/phi3/test_modeling_phi3.py::Phi3IntegrationTest::test_model_phi3_mini_128k_instruct_logits",
874
+ "tests/models/phi3/test_modeling_phi3.py::Phi3IntegrationTest::test_model_phi3_mini_4k_instruct_logits"
875
+ ],
876
+ "multi-gpu": [
877
+ "tests/models/phi3/test_modeling_phi3.py::Phi3IntegrationTest::test_export_static_cache",
878
+ "tests/models/phi3/test_modeling_phi3.py::Phi3IntegrationTest::test_model_phi3_mini_128k_instruct_logits",
879
+ "tests/models/phi3/test_modeling_phi3.py::Phi3IntegrationTest::test_model_phi3_mini_4k_instruct_logits"
880
+ ]
881
+ },
882
+ "phi4_multimodal": {
883
+ "single-gpu": [
884
+ "tests/models/phi4_multimodal/test_modeling_phi4_multimodal.py::Phi4MultimodalIntegrationTest::test_vision_text_generation"
885
+ ],
886
+ "multi-gpu": [
887
+ "tests/models/phi4_multimodal/test_modeling_phi4_multimodal.py::Phi4MultimodalIntegrationTest::test_vision_text_generation"
888
+ ]
889
+ },
890
+ "phimoe": {
891
+ "single-gpu": [
892
+ "tests/models/phimoe/test_modeling_phimoe.py::PhimoeIntegrationTest::test_model_phimoe_instruct_logits",
893
+ "tests/models/phimoe/test_modeling_phimoe.py::PhimoeIntegrationTest::test_phimoe_instruct_generation"
894
+ ],
895
+ "multi-gpu": [
896
+ "tests/models/phimoe/test_modeling_phimoe.py::PhimoeIntegrationTest::test_model_phimoe_instruct_logits",
897
+ "tests/models/phimoe/test_modeling_phimoe.py::PhimoeIntegrationTest::test_phimoe_instruct_generation"
898
+ ]
899
+ },
900
+ "pixtral": {
901
+ "single-gpu": [
902
+ "tests/models/pixtral/test_modeling_pixtral.py::PixtralVisionModelModelTest::test_sdpa_can_dispatch_on_flash"
903
+ ],
904
+ "multi-gpu": [
905
+ "tests/models/pixtral/test_modeling_pixtral.py::PixtralVisionModelModelTest::test_sdpa_can_dispatch_on_flash"
906
+ ]
907
+ },
908
+ "plbart": {
909
+ "single-gpu": [
910
+ "tests/models/plbart/test_modeling_plbart.py::PLBartJavaCsIntegrationTest::test_java_cs_generate_batch",
911
+ "tests/models/plbart/test_modeling_plbart.py::PLBartJavaCsIntegrationTest::test_java_cs_generate_one",
912
+ "tests/models/plbart/test_modeling_plbart.py::PLBartBaseIntegrationTest::test_fill_mask"
913
+ ],
914
+ "multi-gpu": [
915
+ "tests/models/plbart/test_modeling_plbart.py::PLBartJavaCsIntegrationTest::test_java_cs_generate_batch",
916
+ "tests/models/plbart/test_modeling_plbart.py::PLBartJavaCsIntegrationTest::test_java_cs_generate_one",
917
+ "tests/models/plbart/test_modeling_plbart.py::PLBartBaseIntegrationTest::test_fill_mask"
918
+ ]
919
+ },
920
+ "prompt_depth_anything": {
921
+ "single-gpu": [],
922
+ "multi-gpu": [
923
+ "tests/models/prompt_depth_anything/test_modeling_prompt_depth_anything.py::PromptDepthAnythingModelTest::test_model_parallelism"
924
+ ]
925
+ },
926
+ "pvt": {
927
+ "single-gpu": [],
928
+ "multi-gpu": [
929
+ "tests/models/pvt/test_modeling_pvt.py::PvtModelTest::test_model_parallelism"
930
+ ]
931
+ },
932
+ "pvt_v2": {
933
+ "single-gpu": [
934
+ "tests/models/pvt_v2/test_modeling_pvt_v2.py::PvtV2ModelIntegrationTest::test_inference_image_classification",
935
+ "tests/models/pvt_v2/test_modeling_pvt_v2.py::PvtV2ModelIntegrationTest::test_inference_model"
936
+ ],
937
+ "multi-gpu": [
938
+ "tests/models/pvt_v2/test_modeling_pvt_v2.py::PvtV2ModelIntegrationTest::test_inference_image_classification",
939
+ "tests/models/pvt_v2/test_modeling_pvt_v2.py::PvtV2ModelIntegrationTest::test_inference_model"
940
+ ]
941
+ },
942
+ "qwen2": {
943
+ "single-gpu": [
944
+ "tests/models/qwen2/test_modeling_qwen2.py::Qwen2IntegrationTest::test_export_static_cache"
945
+ ],
946
+ "multi-gpu": [
947
+ "tests/models/qwen2/test_modeling_qwen2.py::Qwen2IntegrationTest::test_export_static_cache"
948
+ ]
949
+ },
950
+ "qwen2_5_omni": {
951
+ "single-gpu": [
952
+ "tests/models/qwen2_5_omni/test_modeling_qwen2_5_omni.py::Qwen2_5OmniModelIntegrationTest::test_small_model_integration_test",
953
+ "tests/models/qwen2_5_omni/test_modeling_qwen2_5_omni.py::Qwen2_5OmniModelIntegrationTest::test_small_model_integration_test_batch"
954
+ ],
955
+ "multi-gpu": [
956
+ "tests/models/qwen2_5_omni/test_modeling_qwen2_5_omni.py::Qwen2_5OmniModelIntegrationTest::test_small_model_integration_test",
957
+ "tests/models/qwen2_5_omni/test_modeling_qwen2_5_omni.py::Qwen2_5OmniModelIntegrationTest::test_small_model_integration_test_batch",
958
+ "tests/models/qwen2_5_omni/test_modeling_qwen2_5_omni.py::Qwen2_5OmniThinkerForConditionalGenerationModelTest::test_model_parallelism"
959
+ ]
960
+ },
961
+ "qwen2_moe": {
962
+ "single-gpu": [
963
+ "tests/models/qwen2_moe/test_modeling_qwen2_moe.py::Qwen2MoeIntegrationTest::test_speculative_generation"
964
+ ],
965
+ "multi-gpu": [
966
+ "tests/models/qwen2_moe/test_modeling_qwen2_moe.py::Qwen2MoeIntegrationTest::test_speculative_generation"
967
+ ]
968
+ },
969
+ "qwen2_vl": {
970
+ "single-gpu": [
971
+ "tests/models/qwen2_vl/test_modeling_qwen2_vl.py::Qwen2VLIntegrationTest::test_small_model_integration_test_batch_different_resolutions",
972
+ "tests/models/qwen2_vl/test_video_processing_qwen2_vl.py::Qwen2VLVideoProcessingTest::test_can_compile_fast_video_processor"
973
+ ],
974
+ "multi-gpu": [
975
+ "tests/models/qwen2_vl/test_modeling_qwen2_vl.py::Qwen2VLIntegrationTest::test_small_model_integration_test_batch_different_resolutions",
976
+ "tests/models/qwen2_vl/test_video_processing_qwen2_vl.py::Qwen2VLVideoProcessingTest::test_can_compile_fast_video_processor"
977
+ ]
978
+ },
979
+ "qwen3": {
980
+ "single-gpu": [
981
+ "tests/models/qwen3/test_modeling_qwen3.py::Qwen3IntegrationTest::test_export_static_cache"
982
+ ],
983
+ "multi-gpu": [
984
+ "tests/models/qwen3/test_modeling_qwen3.py::Qwen3IntegrationTest::test_export_static_cache"
985
+ ]
986
+ },
987
+ "qwen3_moe": {
988
+ "single-gpu": [],
989
+ "multi-gpu": [
990
+ "tests/models/qwen3_moe/test_modeling_qwen3_moe.py::Qwen3MoeIntegrationTest::test_model_15b_a2b_generation",
991
+ "tests/models/qwen3_moe/test_modeling_qwen3_moe.py::Qwen3MoeIntegrationTest::test_model_15b_a2b_logits",
992
+ "tests/models/qwen3_moe/test_modeling_qwen3_moe.py::Qwen3MoeIntegrationTest::test_model_15b_a2b_long_prompt_sdpa",
993
+ "tests/models/qwen3_moe/test_modeling_qwen3_moe.py::Qwen3MoeIntegrationTest::test_speculative_generation"
994
+ ]
995
+ },
996
+ "recurrent_gemma": {
997
+ "single-gpu": [
998
+ "tests/models/recurrent_gemma/test_modeling_recurrent_gemma.py::RecurrentGemmaIntegrationTest::test_2b_sample"
999
+ ],
1000
+ "multi-gpu": [
1001
+ "tests/models/recurrent_gemma/test_modeling_recurrent_gemma.py::RecurrentGemmaIntegrationTest::test_2b_sample"
1002
+ ]
1003
+ },
1004
+ "roberta": {
1005
+ "single-gpu": [
1006
+ "tests/models/roberta/test_modeling_roberta.py::RobertaModelTest::test_cpu_offload",
1007
+ "tests/models/roberta/test_modeling_roberta.py::RobertaModelTest::test_disk_offload_bin",
1008
+ "tests/models/roberta/test_modeling_roberta.py::RobertaModelTest::test_disk_offload_safetensors"
1009
+ ],
1010
+ "multi-gpu": [
1011
+ "tests/models/roberta/test_modeling_roberta.py::RobertaModelTest::test_cpu_offload",
1012
+ "tests/models/roberta/test_modeling_roberta.py::RobertaModelTest::test_disk_offload_bin",
1013
+ "tests/models/roberta/test_modeling_roberta.py::RobertaModelTest::test_disk_offload_safetensors",
1014
+ "tests/models/roberta/test_modeling_roberta.py::RobertaModelTest::test_model_parallelism"
1015
+ ]
1016
+ },
1017
+ "sam": {
1018
+ "single-gpu": [
1019
+ "tests/models/sam/test_modeling_sam.py::SamVisionModelTest::test_sdpa_can_dispatch_on_flash"
1020
+ ],
1021
+ "multi-gpu": [
1022
+ "tests/models/sam/test_modeling_sam.py::SamVisionModelTest::test_sdpa_can_dispatch_on_flash",
1023
+ "tests/models/sam/test_modeling_sam.py::SamVisionModelTest::test_model_parallelism",
1024
+ "tests/models/sam/test_modeling_sam.py::SamModelTest::test_model_parallelism"
1025
+ ]
1026
+ },
1027
+ "sam_hq": {
1028
+ "single-gpu": [
1029
+ "tests/models/sam_hq/test_modeling_sam_hq.py::SamHQVisionModelTest::test_sdpa_can_dispatch_on_flash"
1030
+ ],
1031
+ "multi-gpu": [
1032
+ "tests/models/sam_hq/test_modeling_sam_hq.py::SamHQVisionModelTest::test_sdpa_can_dispatch_on_flash",
1033
+ "tests/models/sam_hq/test_modeling_sam_hq.py::SamHQVisionModelTest::test_model_parallelism",
1034
+ "tests/models/sam_hq/test_modeling_sam_hq.py::SamHQModelTest::test_model_parallelism"
1035
+ ]
1036
+ },
1037
+ "seamless_m4t_v2": {
1038
+ "single-gpu": [
1039
+ "tests/models/seamless_m4t_v2/test_modeling_seamless_m4t_v2.py::SeamlessM4Tv2ModelIntegrationTest::test_speech_to_speech_model",
1040
+ "tests/models/seamless_m4t_v2/test_modeling_seamless_m4t_v2.py::SeamlessM4Tv2ModelIntegrationTest::test_text_to_speech_model"
1041
+ ],
1042
+ "multi-gpu": [
1043
+ "tests/models/seamless_m4t_v2/test_modeling_seamless_m4t_v2.py::SeamlessM4Tv2ModelIntegrationTest::test_speech_to_speech_model",
1044
+ "tests/models/seamless_m4t_v2/test_modeling_seamless_m4t_v2.py::SeamlessM4Tv2ModelIntegrationTest::test_text_to_speech_model"
1045
+ ]
1046
+ },
1047
+ "segformer": {
1048
+ "single-gpu": [
1049
+ "tests/models/segformer/test_modeling_segformer.py::SegformerModelTest::test_batching_equivalence"
1050
+ ],
1051
+ "multi-gpu": [
1052
+ "tests/models/segformer/test_modeling_segformer.py::SegformerModelTest::test_batching_equivalence"
1053
+ ]
1054
+ },
1055
+ "shieldgemma2": {
1056
+ "single-gpu": [
1057
+ "tests/models/shieldgemma2/test_modeling_shieldgemma2.py::ShieldGemma2IntegrationTest::test_model"
1058
+ ],
1059
+ "multi-gpu": [
1060
+ "tests/models/shieldgemma2/test_modeling_shieldgemma2.py::ShieldGemma2IntegrationTest::test_model"
1061
+ ]
1062
+ },
1063
+ "speecht5": {
1064
+ "single-gpu": [
1065
+ "tests/models/speecht5/test_modeling_speecht5.py::SpeechT5ForTextToSpeechIntegrationTests::test_batch_generation",
1066
+ "tests/models/speecht5/test_modeling_speecht5.py::SpeechT5ForTextToSpeechIntegrationTests::test_one_to_many_generation"
1067
+ ],
1068
+ "multi-gpu": [
1069
+ "tests/models/speecht5/test_modeling_speecht5.py::SpeechT5ForTextToSpeechIntegrationTests::test_batch_generation",
1070
+ "tests/models/speecht5/test_modeling_speecht5.py::SpeechT5ForTextToSpeechIntegrationTests::test_one_to_many_generation"
1071
+ ]
1072
+ },
1073
+ "superpoint": {
1074
+ "single-gpu": [
1075
+ "tests/models/superpoint/test_modeling_superpoint.py::SuperPointModelIntegrationTest::test_inference"
1076
+ ],
1077
+ "multi-gpu": [
1078
+ "tests/models/superpoint/test_modeling_superpoint.py::SuperPointModelIntegrationTest::test_inference"
1079
+ ]
1080
+ },
1081
+ "swiftformer": {
1082
+ "single-gpu": [
1083
+ "tests/models/swiftformer/test_modeling_swiftformer.py::SwiftFormerModelIntegrationTest::test_inference_image_classification_head"
1084
+ ],
1085
+ "multi-gpu": [
1086
+ "tests/models/swiftformer/test_modeling_swiftformer.py::SwiftFormerModelIntegrationTest::test_inference_image_classification_head"
1087
+ ]
1088
+ },
1089
+ "swin2sr": {
1090
+ "single-gpu": [
1091
+ "tests/models/swin2sr/test_modeling_swin2sr.py::Swin2SRModelIntegrationTest::test_inference_fp16",
1092
+ "tests/models/swin2sr/test_modeling_swin2sr.py::Swin2SRModelIntegrationTest::test_inference_image_super_resolution_head"
1093
+ ],
1094
+ "multi-gpu": [
1095
+ "tests/models/swin2sr/test_modeling_swin2sr.py::Swin2SRModelIntegrationTest::test_inference_fp16",
1096
+ "tests/models/swin2sr/test_modeling_swin2sr.py::Swin2SRModelIntegrationTest::test_inference_image_super_resolution_head"
1097
+ ]
1098
+ },
1099
+ "swinv2": {
1100
+ "single-gpu": [
1101
+ "tests/models/swinv2/test_modeling_swinv2.py::Swinv2ModelIntegrationTest::test_inference_fp16"
1102
+ ],
1103
+ "multi-gpu": [
1104
+ "tests/models/swinv2/test_modeling_swinv2.py::Swinv2ModelIntegrationTest::test_inference_fp16"
1105
+ ]
1106
+ },
1107
+ "t5": {
1108
+ "single-gpu": [
1109
+ "tests/models/t5/test_modeling_t5.py::T5ModelIntegrationTests::test_small_integration_test"
1110
+ ],
1111
+ "multi-gpu": [
1112
+ "tests/models/t5/test_modeling_t5.py::T5ModelIntegrationTests::test_small_integration_test"
1113
+ ]
1114
+ },
1115
+ "table_transformer": {
1116
+ "single-gpu": [
1117
+ "tests/models/table_transformer/test_modeling_table_transformer.py::TableTransformerModelIntegrationTests::test_table_detection"
1118
+ ],
1119
+ "multi-gpu": [
1120
+ "tests/models/table_transformer/test_modeling_table_transformer.py::TableTransformerModelIntegrationTests::test_table_detection"
1121
+ ]
1122
+ },
1123
+ "tapas": {
1124
+ "single-gpu": [
1125
+ "tests/models/tapas/test_modeling_tapas.py::TapasModelIntegrationTest::test_training_question_answering_head_weak_supervision"
1126
+ ],
1127
+ "multi-gpu": [
1128
+ "tests/models/tapas/test_modeling_tapas.py::TapasModelIntegrationTest::test_training_question_answering_head_weak_supervision"
1129
+ ]
1130
+ },
1131
+ "timesfm": {
1132
+ "single-gpu": [],
1133
+ "multi-gpu": [
1134
+ "tests/models/timesfm/test_modeling_timesfm.py::TimesFmModelTest::test_model_parallelism"
1135
+ ]
1136
+ },
1137
+ "udop": {
1138
+ "single-gpu": [
1139
+ "tests/models/udop/test_modeling_udop.py::UdopModelIntegrationTests::test_conditional_generation"
1140
+ ],
1141
+ "multi-gpu": [
1142
+ "tests/models/udop/test_modeling_udop.py::UdopModelIntegrationTests::test_conditional_generation"
1143
+ ]
1144
+ },
1145
+ "umt5": {
1146
+ "single-gpu": [],
1147
+ "multi-gpu": [
1148
+ "tests/models/umt5/test_modeling_umt5.py::UMT5EncoderOnlyModelTest::test_model_parallel_equal_results",
1149
+ "tests/models/umt5/test_modeling_umt5.py::UMT5EncoderOnlyModelTest::test_model_parallelization"
1150
+ ]
1151
+ },
1152
+ "upernet": {
1153
+ "single-gpu": [
1154
+ "tests/models/upernet/test_modeling_upernet.py::UperNetModelIntegrationTest::test_inference_convnext_backbone",
1155
+ "tests/models/upernet/test_modeling_upernet.py::UperNetModelIntegrationTest::test_inference_swin_backbone"
1156
+ ],
1157
+ "multi-gpu": [
1158
+ "tests/models/upernet/test_modeling_upernet.py::UperNetModelIntegrationTest::test_inference_convnext_backbone",
1159
+ "tests/models/upernet/test_modeling_upernet.py::UperNetModelIntegrationTest::test_inference_swin_backbone",
1160
+ "tests/models/upernet/test_modeling_upernet.py::UperNetModelTest::test_model_parallelism"
1161
+ ]
1162
+ },
1163
+ "videomae": {
1164
+ "single-gpu": [
1165
+ "tests/models/videomae/test_modeling_videomae.py::VideoMAEModelIntegrationTest::test_inference_for_pretraining"
1166
+ ],
1167
+ "multi-gpu": [
1168
+ "tests/models/videomae/test_modeling_videomae.py::VideoMAEModelIntegrationTest::test_inference_for_pretraining"
1169
+ ]
1170
+ },
1171
+ "vilt": {
1172
+ "single-gpu": [
1173
+ "tests/models/vilt/test_modeling_vilt.py::ViltModelIntegrationTest::test_inference_natural_language_visual_reasoning"
1174
+ ],
1175
+ "multi-gpu": [
1176
+ "tests/models/vilt/test_modeling_vilt.py::ViltModelIntegrationTest::test_inference_natural_language_visual_reasoning"
1177
+ ]
1178
+ },
1179
+ "vision_encoder_decoder": {
1180
+ "single-gpu": [
1181
+ "tests/models/vision_encoder_decoder/test_modeling_vision_encoder_decoder.py::TrOCRModelIntegrationTest::test_inference_handwritten",
1182
+ "tests/models/vision_encoder_decoder/test_modeling_vision_encoder_decoder.py::TrOCRModelIntegrationTest::test_inference_printed",
1183
+ "tests/models/vision_encoder_decoder/test_modeling_vision_encoder_decoder.py::ViT2GPT2ModelIntegrationTest::test_inference_coco_en"
1184
+ ],
1185
+ "multi-gpu": [
1186
+ "tests/models/vision_encoder_decoder/test_modeling_vision_encoder_decoder.py::TrOCRModelIntegrationTest::test_inference_handwritten",
1187
+ "tests/models/vision_encoder_decoder/test_modeling_vision_encoder_decoder.py::TrOCRModelIntegrationTest::test_inference_printed",
1188
+ "tests/models/vision_encoder_decoder/test_modeling_vision_encoder_decoder.py::ViT2GPT2ModelIntegrationTest::test_inference_coco_en"
1189
+ ]
1190
+ },
1191
+ "vitmatte": {
1192
+ "single-gpu": [
1193
+ "tests/models/vitmatte/test_modeling_vitmatte.py::VitMatteModelTest::test_batching_equivalence"
1194
+ ],
1195
+ "multi-gpu": [
1196
+ "tests/models/vitmatte/test_modeling_vitmatte.py::VitMatteModelTest::test_batching_equivalence"
1197
+ ]
1198
+ },
1199
+ "vitpose": {
1200
+ "single-gpu": [
1201
+ "tests/models/vitpose/test_modeling_vitpose.py::VitPoseModelIntegrationTest::test_batched_inference",
1202
+ "tests/models/vitpose/test_modeling_vitpose.py::VitPoseModelIntegrationTest::test_inference_pose_estimation"
1203
+ ],
1204
+ "multi-gpu": [
1205
+ "tests/models/vitpose/test_modeling_vitpose.py::VitPoseModelIntegrationTest::test_batched_inference",
1206
+ "tests/models/vitpose/test_modeling_vitpose.py::VitPoseModelIntegrationTest::test_inference_pose_estimation"
1207
+ ]
1208
+ },
1209
+ "vits": {
1210
+ "single-gpu": [
1211
+ "tests/models/vits/test_modeling_vits.py::VitsModelIntegrationTests::test_forward_fp16"
1212
+ ],
1213
+ "multi-gpu": [
1214
+ "tests/models/vits/test_modeling_vits.py::VitsModelIntegrationTests::test_forward_fp16"
1215
+ ]
1216
+ },
1217
+ "vivit": {
1218
+ "single-gpu": [],
1219
+ "multi-gpu": [
1220
+ "tests/models/vivit/test_modeling_vivit.py::VivitModelTest::test_model_parallelism"
1221
+ ]
1222
+ },
1223
+ "vjepa2": {
1224
+ "single-gpu": [
1225
+ "tests/models/vjepa2/test_modeling_vjepa2.py::VJEPA2ModelIntegrationTest::test_video_classification"
1226
+ ],
1227
+ "multi-gpu": [
1228
+ "tests/models/vjepa2/test_modeling_vjepa2.py::VJEPA2ModelIntegrationTest::test_video_classification"
1229
+ ]
1230
+ },
1231
+ "xglm": {
1232
+ "single-gpu": [
1233
+ "tests/models/xglm/test_modeling_xglm.py::XGLMModelLanguageGenerationTest::test_batch_generation",
1234
+ "tests/models/xglm/test_modeling_xglm.py::XGLMModelLanguageGenerationTest::test_xglm_sample"
1235
+ ],
1236
+ "multi-gpu": [
1237
+ "tests/models/xglm/test_modeling_xglm.py::XGLMModelLanguageGenerationTest::test_batch_generation",
1238
+ "tests/models/xglm/test_modeling_xglm.py::XGLMModelLanguageGenerationTest::test_xglm_sample"
1239
+ ]
1240
+ },
1241
+ "xlm": {
1242
+ "single-gpu": [
1243
+ "tests/models/xlm/test_modeling_xlm.py::XLMModelLanguageGenerationTest::test_lm_generate_xlm_mlm_en_2048"
1244
+ ],
1245
+ "multi-gpu": [
1246
+ "tests/models/xlm/test_modeling_xlm.py::XLMModelLanguageGenerationTest::test_lm_generate_xlm_mlm_en_2048"
1247
+ ]
1248
+ },
1249
+ "xlnet": {
1250
+ "single-gpu": [
1251
+ "tests/models/xlnet/test_modeling_xlnet.py::XLNetModelLanguageGenerationTest::test_lm_generate_xlnet_base_cased"
1252
+ ],
1253
+ "multi-gpu": [
1254
+ "tests/models/xlnet/test_modeling_xlnet.py::XLNetModelLanguageGenerationTest::test_lm_generate_xlnet_base_cased"
1255
+ ]
1256
+ },
1257
+ "xmod": {
1258
+ "single-gpu": [
1259
+ "tests/models/xmod/test_modeling_xmod.py::XmodModelIntegrationTest::test_end_to_end_mask_fill"
1260
+ ],
1261
+ "multi-gpu": [
1262
+ "tests/models/xmod/test_modeling_xmod.py::XmodModelIntegrationTest::test_end_to_end_mask_fill"
1263
+ ]
1264
+ },
1265
+ "yolos": {
1266
+ "single-gpu": [],
1267
+ "multi-gpu": [
1268
+ "tests/models/yolos/test_modeling_yolos.py::YolosModelTest::test_model_parallelism"
1269
+ ]
1270
+ },
1271
+ "zamba": {
1272
+ "single-gpu": [
1273
+ "tests/models/zamba/test_modeling_zamba.py::ZambaModelIntegrationTest::test_simple_batched_generate_with_padding",
1274
+ "tests/models/zamba/test_modeling_zamba.py::ZambaModelIntegrationTest::test_simple_generate"
1275
+ ],
1276
+ "multi-gpu": [
1277
+ "tests/models/zamba/test_modeling_zamba.py::ZambaModelIntegrationTest::test_simple_batched_generate_with_padding",
1278
+ "tests/models/zamba/test_modeling_zamba.py::ZambaModelIntegrationTest::test_simple_generate"
1279
+ ]
1280
+ },
1281
+ "zamba2": {
1282
+ "single-gpu": [
1283
+ "tests/models/zamba2/test_modeling_zamba2.py::Zamba2ModelTest::test_sdpa_can_dispatch_on_flash",
1284
+ "tests/models/zamba2/test_modeling_zamba2.py::Zamba2ModelIntegrationTest::test_simple_batched_generate_with_padding_0_cuda"
1285
+ ],
1286
+ "multi-gpu": [
1287
+ "tests/models/zamba2/test_modeling_zamba2.py::Zamba2ModelTest::test_sdpa_can_dispatch_on_flash",
1288
+ "tests/models/zamba2/test_modeling_zamba2.py::Zamba2ModelIntegrationTest::test_simple_batched_generate_with_padding_0_cuda"
1289
+ ]
1290
+ },
1291
+ "pipelines": {
1292
+ "single-gpu": [
1293
+ "tests/pipelines/test_pipelines_automatic_speech_recognition.py::AutomaticSpeechRecognitionPipelineTests::test_simple_s2t",
1294
+ "tests/pipelines/test_pipelines_automatic_speech_recognition.py::AutomaticSpeechRecognitionPipelineTests::test_small_model_pt",
1295
+ "tests/pipelines/test_pipelines_automatic_speech_recognition.py::AutomaticSpeechRecognitionPipelineTests::test_small_model_pt_bf16",
1296
+ "tests/pipelines/test_pipelines_automatic_speech_recognition.py::AutomaticSpeechRecognitionPipelineTests::test_small_model_pt_fp16",
1297
+ "tests/pipelines/test_pipelines_common.py::PipelineUtilsTest::test_load_default_pipelines_pt",
1298
+ "tests/pipelines/test_pipelines_common.py::PipelineUtilsTest::test_load_default_pipelines_pt_table_qa",
1299
+ "tests/pipelines/test_pipelines_common.py::CustomPipelineTest::test_cached_pipeline_has_minimum_calls_to_head",
1300
+ "tests/pipelines/test_pipelines_common.py::CustomPipelineTest::test_custom_code_with_string_tokenizer",
1301
+ "tests/pipelines/test_pipelines_depth_estimation.py::DepthEstimationPipelineTests::test_multiprocess",
1302
+ "tests/pipelines/test_pipelines_image_segmentation.py::ImageSegmentationPipelineTests::test_integration_torch_image_segmentation",
1303
+ "tests/pipelines/test_pipelines_image_segmentation.py::ImageSegmentationPipelineTests::test_maskformer",
1304
+ "tests/pipelines/test_pipelines_image_segmentation.py::ImageSegmentationPipelineTests::test_oneformer",
1305
+ "tests/pipelines/test_pipelines_image_text_to_text.py::ImageTextToTextPipelineTests::test_model_pt_chat_template",
1306
+ "tests/pipelines/test_pipelines_question_answering.py::QAPipelineTests::test_large_model_issue",
1307
+ "tests/pipelines/test_pipelines_text_to_audio.py::TextToAudioPipelineTests::test_generative_model_kwargs",
1308
+ "tests/pipelines/test_pipelines_text_to_audio.py::TextToAudioPipelineTests::test_small_musicgen_pt",
1309
+ "tests/pipelines/test_pipelines_token_classification.py::TokenClassificationPipelineTests::test_is_split_into_words",
1310
+ "tests/pipelines/test_pipelines_visual_question_answering.py::VisualQuestionAnsweringPipelineTests::test_large_model_pt_blip2",
1311
+ "tests/pipelines/test_pipelines_visual_question_answering.py::VisualQuestionAnsweringPipelineTests::test_small_model_pt_blip2",
1312
+ "tests/pipelines/test_pipelines_zero_shot_image_classification.py::ZeroShotImageClassificationPipelineTests::test_small_model_pt",
1313
+ "tests/pipelines/test_pipelines_zero_shot_image_classification.py::ZeroShotImageClassificationPipelineTests::test_small_model_pt_fp16"
1314
+ ],
1315
+ "multi-gpu": [
1316
+ "tests/pipelines/test_pipelines_automatic_speech_recognition.py::AutomaticSpeechRecognitionPipelineTests::test_simple_s2t",
1317
+ "tests/pipelines/test_pipelines_automatic_speech_recognition.py::AutomaticSpeechRecognitionPipelineTests::test_small_model_pt",
1318
+ "tests/pipelines/test_pipelines_automatic_speech_recognition.py::AutomaticSpeechRecognitionPipelineTests::test_small_model_pt_bf16",
1319
+ "tests/pipelines/test_pipelines_automatic_speech_recognition.py::AutomaticSpeechRecognitionPipelineTests::test_small_model_pt_fp16",
1320
+ "tests/pipelines/test_pipelines_common.py::PipelineUtilsTest::test_load_default_pipelines_pt",
1321
+ "tests/pipelines/test_pipelines_common.py::PipelineUtilsTest::test_load_default_pipelines_pt_table_qa",
1322
+ "tests/pipelines/test_pipelines_common.py::CustomPipelineTest::test_cached_pipeline_has_minimum_calls_to_head",
1323
+ "tests/pipelines/test_pipelines_common.py::CustomPipelineTest::test_custom_code_with_string_tokenizer",
1324
+ "tests/pipelines/test_pipelines_depth_estimation.py::DepthEstimationPipelineTests::test_multiprocess",
1325
+ "tests/pipelines/test_pipelines_image_segmentation.py::ImageSegmentationPipelineTests::test_integration_torch_image_segmentation",
1326
+ "tests/pipelines/test_pipelines_image_segmentation.py::ImageSegmentationPipelineTests::test_maskformer",
1327
+ "tests/pipelines/test_pipelines_image_segmentation.py::ImageSegmentationPipelineTests::test_oneformer",
1328
+ "tests/pipelines/test_pipelines_image_text_to_text.py::ImageTextToTextPipelineTests::test_model_pt_chat_template",
1329
+ "tests/pipelines/test_pipelines_question_answering.py::QAPipelineTests::test_large_model_issue",
1330
+ "tests/pipelines/test_pipelines_text_to_audio.py::TextToAudioPipelineTests::test_generative_model_kwargs",
1331
+ "tests/pipelines/test_pipelines_text_to_audio.py::TextToAudioPipelineTests::test_small_musicgen_pt",
1332
+ "tests/pipelines/test_pipelines_token_classification.py::TokenClassificationPipelineTests::test_is_split_into_words",
1333
+ "tests/pipelines/test_pipelines_visual_question_answering.py::VisualQuestionAnsweringPipelineTests::test_large_model_pt_blip2",
1334
+ "tests/pipelines/test_pipelines_visual_question_answering.py::VisualQuestionAnsweringPipelineTests::test_small_model_pt_blip2",
1335
+ "tests/pipelines/test_pipelines_zero_shot_image_classification.py::ZeroShotImageClassificationPipelineTests::test_small_model_pt",
1336
+ "tests/pipelines/test_pipelines_zero_shot_image_classification.py::ZeroShotImageClassificationPipelineTests::test_small_model_pt_fp16",
1337
+ "tests/pipelines/test_pipelines_image_segmentation.py::ImageSegmentationPipelineTests::test_threshold"
1338
+ ]
1339
+ },
1340
+ "quantization": {
1341
+ "single-gpu": [
1342
+ "tests/quantization/autoawq/test_awq.py::AwqConfigTest::test_wrong_backend",
1343
+ "tests/quantization/bitnet_integration/test_bitnet.py::BitNetTest::test_quantized_model",
1344
+ "tests/quantization/finegrained_fp8/test_fp8.py::FP8QuantizerTest::test_quantized_model",
1345
+ "tests/quantization/finegrained_fp8/test_fp8.py::FP8QuantizerTest::test_save_pretrained",
1346
+ "tests/quantization/finegrained_fp8/test_fp8.py::FP8QuantizerTest::test_save_pretrained_offload",
1347
+ "tests/quantization/finegrained_fp8/test_fp8.py::FP8LinearTest::test_linear_preserves_shape",
1348
+ "tests/quantization/finegrained_fp8/test_fp8.py::FP8LinearTest::test_linear_with_diff_feature_size_preserves_shape"
1349
+ ],
1350
+ "multi-gpu": [
1351
+ "tests/quantization/autoawq/test_awq.py::AwqConfigTest::test_wrong_backend",
1352
+ "tests/quantization/bitnet_integration/test_bitnet.py::BitNetTest::test_quantized_model",
1353
+ "tests/quantization/finegrained_fp8/test_fp8.py::FP8QuantizerTest::test_quantized_model",
1354
+ "tests/quantization/finegrained_fp8/test_fp8.py::FP8QuantizerTest::test_save_pretrained",
1355
+ "tests/quantization/finegrained_fp8/test_fp8.py::FP8QuantizerTest::test_save_pretrained_offload",
1356
+ "tests/quantization/finegrained_fp8/test_fp8.py::FP8LinearTest::test_linear_preserves_shape",
1357
+ "tests/quantization/finegrained_fp8/test_fp8.py::FP8LinearTest::test_linear_with_diff_feature_size_preserves_shape",
1358
+ "tests/quantization/finegrained_fp8/test_fp8.py::FP8QuantizerTest::test_quantized_model_multi_accelerator",
1359
+ "tests/quantization/finegrained_fp8/test_fp8.py::FP8QuantizerTest::test_save_pretrained_multi_accelerators"
1360
+ ]
1361
+ },
1362
+ "tensor_parallel": {
1363
+ "single-gpu": [
1364
+ "tests/tensor_parallel/test_tensor_parallel.py::TestTensorParallel::test_model_forward",
1365
+ "tests/tensor_parallel/test_tensor_parallel.py::TestTensorParallel::test_model_save"
1366
+ ],
1367
+ "multi-gpu": []
1368
+ },
1369
+ "trainer": {
1370
+ "single-gpu": [
1371
+ "tests/trainer/test_trainer.py::TrainerIntegrationTest::test_auto_batch_size_finder",
1372
+ "tests/trainer/test_trainer.py::TrainerIntegrationTest::test_torchdynamo_memory",
1373
+ "tests/trainer/test_trainer.py::TrainerHyperParameterOptunaIntegrationTest::test_hyperparameter_search",
1374
+ "tests/trainer/test_trainer.py::TrainerHyperParameterMultiObjectOptunaIntegrationTest::test_hyperparameter_search",
1375
+ "tests/trainer/test_trainer.py::TrainerHyperParameterOptunaIntegrationTestWithFullEval::test_hyperparameter_search",
1376
+ "tests/trainer/test_trainer.py::TrainerHyperParameterRayIntegrationTest::test_hyperparameter_search",
1377
+ "tests/trainer/test_trainer.py::TrainerHyperParameterRayIntegrationTest::test_hyperparameter_search_ray_client",
1378
+ "tests/trainer/test_trainer.py::TrainerHyperParameterSigOptIntegrationTest::test_hyperparameter_search"
1379
+ ],
1380
+ "multi-gpu": [
1381
+ "tests/trainer/test_trainer.py::TrainerHyperParameterOptunaIntegrationTest::test_hyperparameter_search",
1382
+ "tests/trainer/test_trainer.py::TrainerHyperParameterMultiObjectOptunaIntegrationTest::test_hyperparameter_search",
1383
+ "tests/trainer/test_trainer.py::TrainerHyperParameterOptunaIntegrationTestWithFullEval::test_hyperparameter_search",
1384
+ "tests/trainer/test_trainer.py::TrainerHyperParameterRayIntegrationTest::test_hyperparameter_search",
1385
+ "tests/trainer/test_trainer.py::TrainerHyperParameterRayIntegrationTest::test_hyperparameter_search_ray_client",
1386
+ "tests/trainer/test_trainer.py::TrainerHyperParameterSigOptIntegrationTest::test_hyperparameter_search",
1387
+ "tests/trainer/test_trainer_fsdp.py::TestFSDPTrainerTorchCompile::test_trainer",
1388
+ "tests/trainer/test_trainer.py::TrainerIntegrationPrerunTest::test_gradient_accumulation",
1389
+ "tests/trainer/test_trainer.py::TrainerIntegrationPrerunTest::test_gradient_accumulation_loss_alignment_with_loss_func",
1390
+ "tests/trainer/test_trainer.py::TrainerIntegrationPrerunTest::test_gradient_accumulation_loss_alignment_with_model_loss",
1391
+ "tests/trainer/test_trainer.py::TrainerIntegrationTest::test_can_resume_training"
1392
+ ]
1393
+ }
1394
+ }