Upload 16 files
Browse files- config.json +112 -164
- generation_config.json +1 -1
- model-00001-of-00004.safetensors +3 -0
- model-00002-of-00004.safetensors +3 -0
- model-00003-of-00004.safetensors +3 -0
- model-00004-of-00004.safetensors +3 -0
- video_preprocessor_config.json +6 -49
config.json
CHANGED
@@ -1,190 +1,138 @@
|
|
1 |
{
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2 |
"architectures": [
|
3 |
-
|
4 |
],
|
5 |
"attention_dropout": 0.0,
|
6 |
"eos_token_id": 151645,
|
7 |
"hidden_act": "silu",
|
8 |
"hidden_size": 3584,
|
9 |
-
"image_token_id":
|
10 |
"initializer_range": 0.02,
|
11 |
"intermediate_size": 18944,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
12 |
"max_position_embeddings": 128000,
|
13 |
"max_window_layers": 28,
|
14 |
-
"model_type": "
|
15 |
"num_attention_heads": 28,
|
16 |
"num_hidden_layers": 28,
|
17 |
"num_key_value_heads": 4,
|
18 |
"pad_token_id": 151654,
|
19 |
"rms_norm_eps": 1e-06,
|
20 |
"rope_scaling": {
|
21 |
-
|
22 |
-
|
23 |
-
|
24 |
-
|
25 |
-
|
26 |
-
|
27 |
-
|
28 |
},
|
29 |
"rope_theta": 1000000.0,
|
30 |
-
"sliding_window":
|
31 |
-
"
|
32 |
-
"architectures": [
|
33 |
-
"Qwen2_5_VLForConditionalGeneration"
|
34 |
-
],
|
35 |
-
"attention_dropout": 0.0,
|
36 |
-
"eos_token_id": 151645,
|
37 |
-
"hidden_act": "silu",
|
38 |
-
"hidden_size": 3584,
|
39 |
-
"image_token_id": null,
|
40 |
-
"initializer_range": 0.02,
|
41 |
-
"intermediate_size": 18944,
|
42 |
-
"max_position_embeddings": 128000,
|
43 |
-
"max_window_layers": 28,
|
44 |
-
"model_type": "qwen2_5_vl_text",
|
45 |
-
"num_attention_heads": 28,
|
46 |
-
"num_hidden_layers": 28,
|
47 |
-
"num_key_value_heads": 4,
|
48 |
-
"pad_token_id": 151654,
|
49 |
-
"quantization_config": {
|
50 |
-
"_load_in_4bit": true,
|
51 |
-
"_load_in_8bit": false,
|
52 |
-
"bnb_4bit_compute_dtype": "bfloat16",
|
53 |
-
"bnb_4bit_quant_storage": "uint8",
|
54 |
-
"bnb_4bit_quant_type": "nf4",
|
55 |
-
"bnb_4bit_use_double_quant": true,
|
56 |
-
"llm_int8_enable_fp32_cpu_offload": false,
|
57 |
-
"llm_int8_has_fp16_weight": false,
|
58 |
-
"llm_int8_skip_modules": [
|
59 |
-
"lm_head",
|
60 |
-
"multi_modal_projector",
|
61 |
-
"merger",
|
62 |
-
"modality_projection",
|
63 |
-
"visual.blocks.27.attn",
|
64 |
-
"visual.blocks.28.attn",
|
65 |
-
"visual.blocks.25.attn",
|
66 |
-
"visual.blocks.22.attn",
|
67 |
-
"visual.blocks.21.attn",
|
68 |
-
"visual.blocks.29.mlp",
|
69 |
-
"visual.blocks.24.attn",
|
70 |
-
"visual.blocks.29.attn",
|
71 |
-
"visual.blocks.31.attn",
|
72 |
-
"visual.blocks.30.attn",
|
73 |
-
"visual.blocks.28.mlp",
|
74 |
-
"visual.blocks.20.attn",
|
75 |
-
"visual.blocks.25.mlp",
|
76 |
-
"visual.blocks.19.attn",
|
77 |
-
"visual.blocks.26.mlp",
|
78 |
-
"visual.blocks.24.mlp",
|
79 |
-
"visual.blocks.17.attn",
|
80 |
-
"visual.blocks.27.mlp",
|
81 |
-
"visual.blocks.23.attn",
|
82 |
-
"visual.blocks.23.mlp",
|
83 |
-
"visual.blocks.21.mlp",
|
84 |
-
"visual.blocks.19.mlp",
|
85 |
-
"visual.blocks.18.attn",
|
86 |
-
"visual.blocks.20.mlp",
|
87 |
-
"visual.blocks.11.attn",
|
88 |
-
"visual.blocks.9.mlp",
|
89 |
-
"visual.blocks.9.attn",
|
90 |
-
"visual.blocks.16.attn",
|
91 |
-
"visual.blocks.11.mlp",
|
92 |
-
"visual.blocks.22.mlp",
|
93 |
-
"visual.blocks.18.mlp",
|
94 |
-
"visual.blocks.13.attn",
|
95 |
-
"visual.blocks.12.attn",
|
96 |
-
"visual.blocks.6.attn",
|
97 |
-
"visual.blocks.10.mlp",
|
98 |
-
"visual.blocks.8.mlp",
|
99 |
-
"visual.blocks.8.attn",
|
100 |
-
"visual.blocks.14.attn",
|
101 |
-
"visual.blocks.4.mlp",
|
102 |
-
"visual.blocks.16.mlp",
|
103 |
-
"visual.blocks.7.mlp",
|
104 |
-
"visual.blocks.6.mlp",
|
105 |
-
"visual.blocks.15.mlp",
|
106 |
-
"visual.blocks.5.mlp",
|
107 |
-
"visual.blocks.10.attn",
|
108 |
-
"visual.blocks.3.mlp",
|
109 |
-
"visual.blocks.12.mlp",
|
110 |
-
"visual.blocks.13.mlp",
|
111 |
-
"visual.blocks.14.mlp",
|
112 |
-
"visual.blocks.2.mlp",
|
113 |
-
"visual.blocks.5.attn",
|
114 |
-
"visual.blocks.1.attn",
|
115 |
-
"visual.blocks.2.attn",
|
116 |
-
"visual.blocks.4.attn",
|
117 |
-
"visual.blocks.3.attn",
|
118 |
-
"visual.blocks.15.attn",
|
119 |
-
"visual.blocks.1.mlp",
|
120 |
-
"visual.blocks.17.mlp",
|
121 |
-
"visual.blocks.0.attn",
|
122 |
-
"visual.blocks.7.attn",
|
123 |
-
"visual.blocks.0.mlp",
|
124 |
-
"visual.blocks.31.mlp.down_proj"
|
125 |
-
],
|
126 |
-
"llm_int8_threshold": 6.0,
|
127 |
-
"load_in_4bit": true,
|
128 |
-
"load_in_8bit": false,
|
129 |
-
"quant_method": "bitsandbytes"
|
130 |
-
},
|
131 |
-
"rms_norm_eps": 1e-06,
|
132 |
-
"rope_scaling": {
|
133 |
-
"mrope_section": [
|
134 |
-
16,
|
135 |
-
24,
|
136 |
-
24
|
137 |
-
],
|
138 |
-
"rope_type": "default",
|
139 |
-
"type": "default"
|
140 |
-
},
|
141 |
-
"rope_theta": 1000000.0,
|
142 |
-
"sliding_window": 32768,
|
143 |
-
"torch_dtype": "float16",
|
144 |
-
"unsloth_fixed": true,
|
145 |
-
"use_cache": true,
|
146 |
-
"use_sliding_window": false,
|
147 |
-
"video_token_id": null,
|
148 |
-
"vision_end_token_id": 151653,
|
149 |
-
"vision_start_token_id": 151652,
|
150 |
-
"vision_token_id": 151654,
|
151 |
-
"vocab_size": 152064
|
152 |
-
},
|
153 |
-
"tie_word_embeddings": false,
|
154 |
-
"torch_dtype": "float16",
|
155 |
-
"transformers_version": "4.52.4",
|
156 |
"unsloth_fixed": true,
|
157 |
-
"unsloth_version": "2025.6.8",
|
158 |
"use_cache": true,
|
159 |
"use_sliding_window": false,
|
160 |
-
"video_token_id":
|
161 |
-
"vision_config": {
|
162 |
-
"depth": 32,
|
163 |
-
"fullatt_block_indexes": [
|
164 |
-
7,
|
165 |
-
15,
|
166 |
-
23,
|
167 |
-
31
|
168 |
-
],
|
169 |
-
"hidden_act": "silu",
|
170 |
-
"hidden_size": 1280,
|
171 |
-
"in_channels": 3,
|
172 |
-
"in_chans": 3,
|
173 |
-
"initializer_range": 0.02,
|
174 |
-
"intermediate_size": 3420,
|
175 |
-
"model_type": "qwen2_5_vl",
|
176 |
-
"num_heads": 16,
|
177 |
-
"out_hidden_size": 3584,
|
178 |
-
"patch_size": 14,
|
179 |
-
"spatial_merge_size": 2,
|
180 |
-
"spatial_patch_size": 14,
|
181 |
-
"temporal_patch_size": 2,
|
182 |
-
"tokens_per_second": 2,
|
183 |
-
"torch_dtype": "float16",
|
184 |
-
"window_size": 112
|
185 |
-
},
|
186 |
"vision_end_token_id": 151653,
|
187 |
"vision_start_token_id": 151652,
|
188 |
"vision_token_id": 151654,
|
189 |
"vocab_size": 152064
|
190 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
{
|
2 |
+
"architectures": [
|
3 |
+
"Qwen2_5_VLForConditionalGeneration"
|
4 |
+
],
|
5 |
+
"attention_dropout": 0.0,
|
6 |
+
"eos_token_id": 151645,
|
7 |
+
"hidden_act": "silu",
|
8 |
+
"hidden_size": 3584,
|
9 |
+
"image_token_id": 151655,
|
10 |
+
"initializer_range": 0.02,
|
11 |
+
"intermediate_size": 18944,
|
12 |
+
"max_position_embeddings": 128000,
|
13 |
+
"max_window_layers": 28,
|
14 |
+
"model_type": "qwen2_5_vl",
|
15 |
+
"num_attention_heads": 28,
|
16 |
+
"num_hidden_layers": 28,
|
17 |
+
"num_key_value_heads": 4,
|
18 |
+
"pad_token_id": 151654,
|
19 |
+
"rms_norm_eps": 1e-06,
|
20 |
+
"rope_scaling": {
|
21 |
+
"mrope_section": [
|
22 |
+
16,
|
23 |
+
24,
|
24 |
+
24
|
25 |
+
],
|
26 |
+
"rope_type": "default",
|
27 |
+
"type": "default"
|
28 |
+
},
|
29 |
+
"rope_theta": 1000000.0,
|
30 |
+
"sliding_window": 32768,
|
31 |
+
"text_config": {
|
32 |
"architectures": [
|
33 |
+
"Qwen2_5_VLForConditionalGeneration"
|
34 |
],
|
35 |
"attention_dropout": 0.0,
|
36 |
"eos_token_id": 151645,
|
37 |
"hidden_act": "silu",
|
38 |
"hidden_size": 3584,
|
39 |
+
"image_token_id": null,
|
40 |
"initializer_range": 0.02,
|
41 |
"intermediate_size": 18944,
|
42 |
+
"layer_types": [
|
43 |
+
"full_attention",
|
44 |
+
"full_attention",
|
45 |
+
"full_attention",
|
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 |
+
],
|
72 |
"max_position_embeddings": 128000,
|
73 |
"max_window_layers": 28,
|
74 |
+
"model_type": "qwen2_5_vl_text",
|
75 |
"num_attention_heads": 28,
|
76 |
"num_hidden_layers": 28,
|
77 |
"num_key_value_heads": 4,
|
78 |
"pad_token_id": 151654,
|
79 |
"rms_norm_eps": 1e-06,
|
80 |
"rope_scaling": {
|
81 |
+
"mrope_section": [
|
82 |
+
16,
|
83 |
+
24,
|
84 |
+
24
|
85 |
+
],
|
86 |
+
"rope_type": "default",
|
87 |
+
"type": "default"
|
88 |
},
|
89 |
"rope_theta": 1000000.0,
|
90 |
+
"sliding_window": null,
|
91 |
+
"torch_dtype": "bfloat16",
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
92 |
"unsloth_fixed": true,
|
|
|
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 |
+
"torch_dtype": "bfloat16",
|
103 |
+
"transformers_version": "4.53.0",
|
104 |
+
"unsloth_fixed": true,
|
105 |
+
"unsloth_version": "2025.6.12",
|
106 |
+
"use_cache": true,
|
107 |
+
"use_sliding_window": false,
|
108 |
+
"video_token_id": 151656,
|
109 |
+
"vision_config": {
|
110 |
+
"depth": 32,
|
111 |
+
"fullatt_block_indexes": [
|
112 |
+
7,
|
113 |
+
15,
|
114 |
+
23,
|
115 |
+
31
|
116 |
+
],
|
117 |
+
"hidden_act": "silu",
|
118 |
+
"hidden_size": 1280,
|
119 |
+
"in_channels": 3,
|
120 |
+
"in_chans": 3,
|
121 |
+
"initializer_range": 0.02,
|
122 |
+
"intermediate_size": 3420,
|
123 |
+
"model_type": "qwen2_5_vl",
|
124 |
+
"num_heads": 16,
|
125 |
+
"out_hidden_size": 3584,
|
126 |
+
"patch_size": 14,
|
127 |
+
"spatial_merge_size": 2,
|
128 |
+
"spatial_patch_size": 14,
|
129 |
+
"temporal_patch_size": 2,
|
130 |
+
"tokens_per_second": 2,
|
131 |
+
"torch_dtype": "bfloat16",
|
132 |
+
"window_size": 112
|
133 |
+
},
|
134 |
+
"vision_end_token_id": 151653,
|
135 |
+
"vision_start_token_id": 151652,
|
136 |
+
"vision_token_id": 151654,
|
137 |
+
"vocab_size": 152064
|
138 |
+
}
|
generation_config.json
CHANGED
@@ -9,5 +9,5 @@
|
|
9 |
"pad_token_id": 151654,
|
10 |
"repetition_penalty": 1.05,
|
11 |
"temperature": 1e-06,
|
12 |
-
"transformers_version": "4.
|
13 |
}
|
|
|
9 |
"pad_token_id": 151654,
|
10 |
"repetition_penalty": 1.05,
|
11 |
"temperature": 1e-06,
|
12 |
+
"transformers_version": "4.53.0"
|
13 |
}
|
model-00001-of-00004.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:d725335e4ea2399be706469e4b8807716a8fa64bd03468252e9f7acf2415fee4
|
3 |
+
size 4968243304
|
model-00002-of-00004.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:b1830db6908dcc76df3a71492acbcf2b8cac130114cf1f3c2d9edae8de8c6de3
|
3 |
+
size 4991495816
|
model-00003-of-00004.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:09c1807c6d00d7cab94f7db39d4c02ebb8537225ccde383861ac48db97945aa6
|
3 |
+
size 4932751040
|
model-00004-of-00004.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:5dd068336d14d45ffb43cef374d286cc6ba9d8741b028f90a7d040d847961f4a
|
3 |
+
size 1691924384
|
video_preprocessor_config.json
CHANGED
@@ -1,28 +1,4 @@
|
|
1 |
{
|
2 |
-
"_valid_kwargs_names": [
|
3 |
-
"do_convert_rgb",
|
4 |
-
"do_resize",
|
5 |
-
"size",
|
6 |
-
"size_divisor",
|
7 |
-
"default_to_square",
|
8 |
-
"resample",
|
9 |
-
"do_rescale",
|
10 |
-
"rescale_factor",
|
11 |
-
"do_normalize",
|
12 |
-
"image_mean",
|
13 |
-
"image_std",
|
14 |
-
"do_pad",
|
15 |
-
"do_center_crop",
|
16 |
-
"crop_size",
|
17 |
-
"data_format",
|
18 |
-
"input_data_format",
|
19 |
-
"device",
|
20 |
-
"min_pixels",
|
21 |
-
"max_pixels",
|
22 |
-
"patch_size",
|
23 |
-
"temporal_patch_size",
|
24 |
-
"merge_size"
|
25 |
-
],
|
26 |
"crop_size": null,
|
27 |
"data_format": "channels_first",
|
28 |
"default_to_square": true,
|
@@ -33,45 +9,25 @@
|
|
33 |
"do_pad": null,
|
34 |
"do_rescale": true,
|
35 |
"do_resize": true,
|
|
|
|
|
36 |
"image_mean": [
|
37 |
0.48145466,
|
38 |
0.4578275,
|
39 |
0.40821073
|
40 |
],
|
41 |
-
"image_processor_type": "Qwen2VLImageProcessor",
|
42 |
"image_std": [
|
43 |
0.26862954,
|
44 |
0.26130258,
|
45 |
0.27577711
|
46 |
],
|
47 |
"input_data_format": null,
|
|
|
48 |
"max_pixels": 12845056,
|
49 |
"merge_size": 2,
|
|
|
50 |
"min_pixels": 3136,
|
51 |
-
"
|
52 |
-
"do_convert_rgb",
|
53 |
-
"do_resize",
|
54 |
-
"size",
|
55 |
-
"size_divisor",
|
56 |
-
"default_to_square",
|
57 |
-
"resample",
|
58 |
-
"do_rescale",
|
59 |
-
"rescale_factor",
|
60 |
-
"do_normalize",
|
61 |
-
"image_mean",
|
62 |
-
"image_std",
|
63 |
-
"do_pad",
|
64 |
-
"do_center_crop",
|
65 |
-
"crop_size",
|
66 |
-
"data_format",
|
67 |
-
"input_data_format",
|
68 |
-
"device",
|
69 |
-
"min_pixels",
|
70 |
-
"max_pixels",
|
71 |
-
"patch_size",
|
72 |
-
"temporal_patch_size",
|
73 |
-
"merge_size"
|
74 |
-
],
|
75 |
"patch_size": 14,
|
76 |
"processor_class": "Qwen2_5_VLProcessor",
|
77 |
"resample": 3,
|
@@ -82,5 +38,6 @@
|
|
82 |
},
|
83 |
"size_divisor": null,
|
84 |
"temporal_patch_size": 2,
|
|
|
85 |
"video_processor_type": "Qwen2VLVideoProcessor"
|
86 |
}
|
|
|
1 |
{
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2 |
"crop_size": null,
|
3 |
"data_format": "channels_first",
|
4 |
"default_to_square": true,
|
|
|
9 |
"do_pad": null,
|
10 |
"do_rescale": true,
|
11 |
"do_resize": true,
|
12 |
+
"do_sample_frames": false,
|
13 |
+
"fps": null,
|
14 |
"image_mean": [
|
15 |
0.48145466,
|
16 |
0.4578275,
|
17 |
0.40821073
|
18 |
],
|
|
|
19 |
"image_std": [
|
20 |
0.26862954,
|
21 |
0.26130258,
|
22 |
0.27577711
|
23 |
],
|
24 |
"input_data_format": null,
|
25 |
+
"max_frames": 768,
|
26 |
"max_pixels": 12845056,
|
27 |
"merge_size": 2,
|
28 |
+
"min_frames": 4,
|
29 |
"min_pixels": 3136,
|
30 |
+
"num_frames": null,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
31 |
"patch_size": 14,
|
32 |
"processor_class": "Qwen2_5_VLProcessor",
|
33 |
"resample": 3,
|
|
|
38 |
},
|
39 |
"size_divisor": null,
|
40 |
"temporal_patch_size": 2,
|
41 |
+
"video_metadata": null,
|
42 |
"video_processor_type": "Qwen2VLVideoProcessor"
|
43 |
}
|