Hugging Face's logo

SoybeanMilk
/
Kimi-VL-A3B-Thinking-2506-BNB-4bit

Image-Text-to-Text
Transformers
Safetensors
kimi_vl
feature-extraction
conversational
custom_code
4-bit precision
bitsandbytes
Model card Files
xet
Community
1
SoybeanMilk commited on
Commit
10e582a
·
verified ·
1 Parent(s): 5d4437b

Upload folder using huggingface_hub

Browse files
Files changed (9) hide show
  1. chat_template.jinja +31 -0
  2. config.json +92 -0
  3. generation_config.json +10 -0
  4. model-00001-of-00002.safetensors +3 -0
  5. model-00002-of-00002.safetensors +3 -0
  6. model.safetensors.index.json +0 -0
  7. special_tokens_map.json +41 -0
  8. tiktoken.model +3 -0
  9. tokenizer_config.json +133 -0
chat_template.jinja ADDED
@@ -0,0 +1,31 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {%- for message in messages -%}
2
+ {%- if loop.first and messages[0]['role'] != 'system' -%}
3
+ {{'<|im_system|>system<|im_middle|>You are a helpful assistant<|im_end|>'}}
4
+ {%- endif -%}
5
+ {%- if message['role'] == 'system' -%}
6
+ {{'<|im_system|>'}}
7
+ {%- endif -%}
8
+ {%- if message['role'] == 'user' -%}
9
+ {{'<|im_user|>'}}
10
+ {%- endif -%}
11
+ {%- if message['role'] == 'assistant' -%}
12
+ {{'<|im_assistant|>'}}
13
+ {%- endif -%}
14
+ {{- message['role'] -}}
15
+ {{'<|im_middle|>'}}
16
+ {%- if message['content'] is string -%}
17
+ {{- message['content'] + '<|im_end|>' -}}
18
+ {%- else -%}
19
+ {%- for content in message['content'] -%}
20
+ {%- if content['type'] == 'image' or 'image' in content or 'image_url' in content -%}
21
+ {{'<|media_start|>image<|media_content|><|media_pad|><|media_end|>'}}
22
+ {%- else -%}
23
+ {{content['text']}}
24
+ {%- endif -%}
25
+ {%- endfor -%}
26
+ {{'<|im_end|>'}}
27
+ {%- endif -%}
28
+ {%- endfor -%}
29
+ {%- if add_generation_prompt -%}
30
+ {{'<|im_assistant|>assistant<|im_middle|>'}}
31
+ {%- endif -%}
config.json ADDED
@@ -0,0 +1,92 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "KimiVLForConditionalGeneration"
4
+ ],
5
+ "auto_map": {
6
+ "AutoConfig": "moonshotai/Kimi-VL-A3B-Thinking-2506--configuration_kimi_vl.KimiVLConfig",
7
+ "AutoModel": "moonshotai/Kimi-VL-A3B-Thinking-2506--modeling_kimi_vl.KimiVLForConditionalGeneration",
8
+ "AutoModelForCausalLM": "moonshotai/Kimi-VL-A3B-Thinking-2506--modeling_kimi_vl.KimiVLForConditionalGeneration"
9
+ },
10
+ "ignore_index": -100,
11
+ "media_placeholder_token_id": 163605,
12
+ "model_type": "kimi_vl",
13
+ "pad_token_id": 0,
14
+ "quantization_config": {
15
+ "_load_in_4bit": true,
16
+ "_load_in_8bit": false,
17
+ "bnb_4bit_compute_dtype": "float16",
18
+ "bnb_4bit_quant_storage": "uint8",
19
+ "bnb_4bit_quant_type": "nf4",
20
+ "bnb_4bit_use_double_quant": true,
21
+ "llm_int8_enable_fp32_cpu_offload": false,
22
+ "llm_int8_has_fp16_weight": false,
23
+ "llm_int8_skip_modules": null,
24
+ "llm_int8_threshold": 6.0,
25
+ "load_in_4bit": true,
26
+ "load_in_8bit": false,
27
+ "quant_method": "bitsandbytes"
28
+ },
29
+ "text_config": {
30
+ "attention_bias": false,
31
+ "attention_dropout": 0.0,
32
+ "aux_loss_alpha": 0.001,
33
+ "bos_token_id": 163584,
34
+ "eos_token_id": 163585,
35
+ "ep_size": 1,
36
+ "first_k_dense_replace": 1,
37
+ "hidden_act": "silu",
38
+ "hidden_size": 2048,
39
+ "initializer_range": 0.02,
40
+ "intermediate_size": 11264,
41
+ "kv_lora_rank": 512,
42
+ "max_position_embeddings": 131072,
43
+ "model_type": "deepseek_v3",
44
+ "moe_intermediate_size": 1408,
45
+ "moe_layer_freq": 1,
46
+ "n_group": 1,
47
+ "n_routed_experts": 64,
48
+ "n_shared_experts": 2,
49
+ "norm_topk_prob": true,
50
+ "num_attention_heads": 16,
51
+ "num_experts_per_tok": 6,
52
+ "num_hidden_layers": 27,
53
+ "num_key_value_heads": 16,
54
+ "num_nextn_predict_layers": 1,
55
+ "pad_token_id": 163839,
56
+ "pretraining_tp": 1,
57
+ "q_lora_rank": null,
58
+ "qk_nope_head_dim": 128,
59
+ "qk_rope_head_dim": 64,
60
+ "rms_norm_eps": 1e-05,
61
+ "rope_scaling": null,
62
+ "rope_theta": 800000.0,
63
+ "routed_scaling_factor": 2.446,
64
+ "scoring_func": "sigmoid",
65
+ "seq_aux": true,
66
+ "topk_group": 1,
67
+ "topk_method": "noaux_tc",
68
+ "torch_dtype": "bfloat16",
69
+ "use_cache": true,
70
+ "v_head_dim": 128,
71
+ "vocab_size": 163840
72
+ },
73
+ "tie_word_embeddings": false,
74
+ "torch_dtype": "float16",
75
+ "transformers_version": "4.52.4",
76
+ "vision_config": {
77
+ "hidden_size": 1152,
78
+ "init_pos_emb_height": 64,
79
+ "init_pos_emb_width": 64,
80
+ "intermediate_size": 4304,
81
+ "merge_kernel_size": [
82
+ 2,
83
+ 2
84
+ ],
85
+ "model_type": "moonvit",
86
+ "num_attention_heads": 16,
87
+ "num_hidden_layers": 27,
88
+ "patch_size": 14,
89
+ "torch_dtype": "bfloat16"
90
+ },
91
+ "vocab_size": 163840
92
+ }
generation_config.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token_id": 163584,
3
+ "do_sample": true,
4
+ "eos_token_id": [
5
+ 163585
6
+ ],
7
+ "pad_token_id": 163838,
8
+ "temperature": 0.6,
9
+ "transformers_version": "4.52.4"
10
+ }
model-00001-of-00002.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d91b6675095f1e591b0c7c53d02f925c152353403d44266a8671bcf512e4198c
3
+ size 5001262096
model-00002-of-00002.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ba007c5c6f266f34e6f637bdc0a1c4ea366a8739a4cf82175f63e03c03f4086d
3
+ size 4484025285
model.safetensors.index.json ADDED
The diff for this file is too large to render. See raw diff
 
special_tokens_map.json ADDED
@@ -0,0 +1,41 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ "<|im_end|>",
4
+ "<|im_user|>",
5
+ "<|im_assistant|>",
6
+ "<|im_system|>",
7
+ "<|im_middle|>",
8
+ "<|media_start|>",
9
+ "<|media_content|>",
10
+ "<|media_end|>",
11
+ "<|media_pad|>"
12
+ ],
13
+ "bos_token": {
14
+ "content": "[BOS]",
15
+ "lstrip": false,
16
+ "normalized": false,
17
+ "rstrip": false,
18
+ "single_word": false
19
+ },
20
+ "eos_token": {
21
+ "content": "[EOS]",
22
+ "lstrip": false,
23
+ "normalized": false,
24
+ "rstrip": false,
25
+ "single_word": false
26
+ },
27
+ "pad_token": {
28
+ "content": "[PAD]",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false
33
+ },
34
+ "unk_token": {
35
+ "content": "[UNK]",
36
+ "lstrip": false,
37
+ "normalized": false,
38
+ "rstrip": false,
39
+ "single_word": false
40
+ }
41
+ }
tiktoken.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b6c497a7469b33ced9c38afb1ad6e47f03f5e5dc05f15930799210ec050c5103
3
+ size 2795286
tokenizer_config.json ADDED
@@ -0,0 +1,133 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "163584": {
4
+ "content": "[BOS]",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "163585": {
12
+ "content": "[EOS]",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "163586": {
20
+ "content": "<|im_end|>",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "163587": {
28
+ "content": "<|im_user|>",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "163588": {
36
+ "content": "<|im_assistant|>",
37
+ "lstrip": false,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ },
43
+ "163594": {
44
+ "content": "<|im_system|>",
45
+ "lstrip": false,
46
+ "normalized": false,
47
+ "rstrip": false,
48
+ "single_word": false,
49
+ "special": true
50
+ },
51
+ "163601": {
52
+ "content": "<|im_middle|>",
53
+ "lstrip": false,
54
+ "normalized": false,
55
+ "rstrip": false,
56
+ "single_word": false,
57
+ "special": true
58
+ },
59
+ "163602": {
60
+ "content": "<|media_start|>",
61
+ "lstrip": false,
62
+ "normalized": false,
63
+ "rstrip": false,
64
+ "single_word": false,
65
+ "special": true
66
+ },
67
+ "163603": {
68
+ "content": "<|media_content|>",
69
+ "lstrip": false,
70
+ "normalized": false,
71
+ "rstrip": false,
72
+ "single_word": false,
73
+ "special": true
74
+ },
75
+ "163604": {
76
+ "content": "<|media_end|>",
77
+ "lstrip": false,
78
+ "normalized": false,
79
+ "rstrip": false,
80
+ "single_word": false,
81
+ "special": true
82
+ },
83
+ "163605": {
84
+ "content": "<|media_pad|>",
85
+ "lstrip": false,
86
+ "normalized": false,
87
+ "rstrip": false,
88
+ "single_word": false,
89
+ "special": true
90
+ },
91
+ "163838": {
92
+ "content": "[PAD]",
93
+ "lstrip": false,
94
+ "normalized": false,
95
+ "rstrip": false,
96
+ "single_word": false,
97
+ "special": true
98
+ },
99
+ "163839": {
100
+ "content": "[UNK]",
101
+ "lstrip": false,
102
+ "normalized": false,
103
+ "rstrip": false,
104
+ "single_word": false,
105
+ "special": true
106
+ }
107
+ },
108
+ "additional_special_tokens": [
109
+ "<|im_end|>",
110
+ "<|im_user|>",
111
+ "<|im_assistant|>",
112
+ "<|im_system|>",
113
+ "<|im_middle|>",
114
+ "<|media_start|>",
115
+ "<|media_content|>",
116
+ "<|media_end|>",
117
+ "<|media_pad|>"
118
+ ],
119
+ "auto_map": {
120
+ "AutoTokenizer": [
121
+ "moonshotai/Kimi-VL-A3B-Thinking-2506--tokenization_moonshot.TikTokenTokenizer",
122
+ null
123
+ ]
124
+ },
125
+ "bos_token": "[BOS]",
126
+ "clean_up_tokenization_spaces": false,
127
+ "eos_token": "[EOS]",
128
+ "extra_special_tokens": {},
129
+ "model_max_length": 1048576,
130
+ "pad_token": "[PAD]",
131
+ "tokenizer_class": "TikTokenTokenizer",
132
+ "unk_token": "[UNK]"
133
+ }