Hugging Face's logo

dahwinsingularity
/
vllm

Image-Text-to-Text
Transformers
Safetensors
qwen2_5_vl
conversational
text-generation-inference
Model card Files
xet
 Community
dahwinsingularity commited on
Commit
6f389f5
·
verified ·
1 Parent(s): 8e15cea

Upload Qwen2_5_VLForConditionalGeneration

Browse files
Files changed (6) hide show
  1. config.json +98 -14
  2. generation_config.json +2 -4
  3. model-00001-of-00003.safetensors +3 -0
  4. model-00002-of-00003.safetensors +3 -0
  5. model-00003-of-00003.safetensors +3 -0
  6. model.safetensors.index.json +0 -0
config.json CHANGED
@@ -1,22 +1,22 @@
1
  {
2
- "_name_or_path": "Qwen/Qwen2.5-VL-3B-Instruct",
3
  "architectures": [
4
  "Qwen2_5_VLForConditionalGeneration"
5
  ],
6
  "attention_dropout": 0.0,
7
  "bos_token_id": 151643,
 
8
  "eos_token_id": 151645,
9
  "hidden_act": "silu",
10
- "hidden_size": 2048,
11
  "image_token_id": 151655,
12
  "initializer_range": 0.02,
13
- "intermediate_size": 11008,
14
  "max_position_embeddings": 128000,
15
- "max_window_layers": 70,
16
  "model_type": "qwen2_5_vl",
17
- "num_attention_heads": 16,
18
- "num_hidden_layers": 36,
19
- "num_key_value_heads": 2,
20
  "rms_norm_eps": 1e-06,
21
  "rope_scaling": {
22
  "mrope_section": [
@@ -29,23 +29,107 @@
29
  },
30
  "rope_theta": 1000000.0,
31
  "sliding_window": 32768,
32
- "tie_word_embeddings": true,
33
- "torch_dtype": "bfloat16",
34
- "transformers_version": "4.49.0.dev0",
35
- "use_cache": true,
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
36
  "use_sliding_window": false,
37
  "video_token_id": 151656,
38
  "vision_config": {
 
 
 
 
 
 
 
 
 
39
  "hidden_size": 1280,
 
40
  "in_chans": 3,
 
 
41
  "model_type": "qwen2_5_vl",
42
- "out_hidden_size": 2048,
 
 
 
43
  "spatial_patch_size": 14,
 
44
  "tokens_per_second": 2,
45
- "torch_dtype": "bfloat16"
46
  },
47
  "vision_end_token_id": 151653,
48
  "vision_start_token_id": 151652,
49
  "vision_token_id": 151654,
50
- "vocab_size": 151936
51
  }
 
1
  {
 
2
  "architectures": [
3
  "Qwen2_5_VLForConditionalGeneration"
4
  ],
5
  "attention_dropout": 0.0,
6
  "bos_token_id": 151643,
7
+ "dtype": "bfloat16",
8
  "eos_token_id": 151645,
9
  "hidden_act": "silu",
10
+ "hidden_size": 3584,
11
  "image_token_id": 151655,
12
  "initializer_range": 0.02,
13
+ "intermediate_size": 18944,
14
  "max_position_embeddings": 128000,
15
+ "max_window_layers": 28,
16
  "model_type": "qwen2_5_vl",
17
+ "num_attention_heads": 28,
18
+ "num_hidden_layers": 28,
19
+ "num_key_value_heads": 4,
20
  "rms_norm_eps": 1e-06,
21
  "rope_scaling": {
22
  "mrope_section": [
 
29
  },
30
  "rope_theta": 1000000.0,
31
  "sliding_window": 32768,
32
+ "text_config": {
33
+ "architectures": [
34
+ "Qwen2_5_VLForConditionalGeneration"
35
+ ],
36
+ "attention_dropout": 0.0,
37
+ "bos_token_id": 151643,
38
+ "dtype": "bfloat16",
39
+ "eos_token_id": 151645,
40
+ "hidden_act": "silu",
41
+ "hidden_size": 3584,
42
+ "image_token_id": null,
43
+ "initializer_range": 0.02,
44
+ "intermediate_size": 18944,
45
+ "layer_types": [
46
+ "full_attention",
47
+ "full_attention",
48
+ "full_attention",
49
+ "full_attention",
50
+ "full_attention",
51
+ "full_attention",
52
+ "full_attention",
53
+ "full_attention",
54
+ "full_attention",
55
+ "full_attention",
56
+ "full_attention",
57
+ "full_attention",
58
+ "full_attention",
59
+ "full_attention",
60
+ "full_attention",
61
+ "full_attention",
62
+ "full_attention",
63
+ "full_attention",
64
+ "full_attention",
65
+ "full_attention",
66
+ "full_attention",
67
+ "full_attention",
68
+ "full_attention",
69
+ "full_attention",
70
+ "full_attention",
71
+ "full_attention",
72
+ "full_attention",
73
+ "full_attention"
74
+ ],
75
+ "max_position_embeddings": 128000,
76
+ "max_window_layers": 28,
77
+ "model_type": "qwen2_5_vl_text",
78
+ "num_attention_heads": 28,
79
+ "num_hidden_layers": 28,
80
+ "num_key_value_heads": 4,
81
+ "rms_norm_eps": 1e-06,
82
+ "rope_scaling": {
83
+ "mrope_section": [
84
+ 16,
85
+ 24,
86
+ 24
87
+ ],
88
+ "rope_type": "default",
89
+ "type": "default"
90
+ },
91
+ "rope_theta": 1000000.0,
92
+ "sliding_window": null,
93
+ "use_cache": true,
94
+ "use_sliding_window": false,
95
+ "video_token_id": null,
96
+ "vision_end_token_id": 151653,
97
+ "vision_start_token_id": 151652,
98
+ "vision_token_id": 151654,
99
+ "vocab_size": 152064
100
+ },
101
+ "tie_word_embeddings": false,
102
+ "transformers_version": "4.56.1",
103
+ "use_cache": false,
104
  "use_sliding_window": false,
105
  "video_token_id": 151656,
106
  "vision_config": {
107
+ "depth": 32,
108
+ "dtype": "bfloat16",
109
+ "fullatt_block_indexes": [
110
+ 7,
111
+ 15,
112
+ 23,
113
+ 31
114
+ ],
115
+ "hidden_act": "silu",
116
  "hidden_size": 1280,
117
+ "in_channels": 3,
118
  "in_chans": 3,
119
+ "initializer_range": 0.02,
120
+ "intermediate_size": 3420,
121
  "model_type": "qwen2_5_vl",
122
+ "num_heads": 16,
123
+ "out_hidden_size": 3584,
124
+ "patch_size": 14,
125
+ "spatial_merge_size": 2,
126
  "spatial_patch_size": 14,
127
+ "temporal_patch_size": 2,
128
  "tokens_per_second": 2,
129
+ "window_size": 112
130
  },
131
  "vision_end_token_id": 151653,
132
  "vision_start_token_id": 151652,
133
  "vision_token_id": 151654,
134
+ "vocab_size": 152064
135
  }
generation_config.json CHANGED
@@ -7,8 +7,6 @@
7
  ],
8
  "pad_token_id": 151643,
9
  "repetition_penalty": 1.05,
10
- "temperature": 0.1,
11
- "top_k": 1,
12
- "top_p": 0.001,
13
- "transformers_version": "4.49.0.dev0"
14
  }
 
7
  ],
8
  "pad_token_id": 151643,
9
  "repetition_penalty": 1.05,
10
+ "temperature": 1e-06,
11
+ "transformers_version": "4.56.1"
 
 
12
  }
model-00001-of-00003.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ac6f80df9b2f1d39fa60eda17bcbe10b4bd97b56de25ad8c87155b7ecfa7af1f
3
+ size 4965216032
model-00002-of-00003.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:94afcd44270fc1167c3f632628f9241a529a6339e6c0098c876427d00c875047
3
+ size 4932751008
model-00003-of-00003.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6fc50a63a631da297de7a25788d2bdcaf3e74bd22d9f329e49927bcee0f703ab
3
+ size 4330857944
model.safetensors.index.json CHANGED
The diff for this file is too large to render. See raw diff