Hugging Face's logo

KennyVo
/
KAT-Coder-V2.5-Dev-OrnithVision

Image-Text-to-Text
Safetensors
qwen3_5_moe
vision
thinking
w4a16
awq
int4
compressed-tensors
conversational
Model card Files
xet
 Community
KennyVo commited on
Commit
fb4db7a
·
verified ·
1 Parent(s): d4e57d2

Upload KAT-Coder-V2.5-Dev with Ornith visual tower

Browse files

Transplanted Ornith-1.5-35B-A3B visual tower weights into KAT-Coder-V2.5-Dev. All 333 model.visual.* tensors replaced. Language model weights preserved.

Files changed (11) hide show
  1. .gitattributes +2 -0
  2. config.json +718 -0
  3. model-00001-of-00005.safetensors +3 -0
  4. model-00002-of-00005.safetensors +3 -0
  5. model-00003-of-00005.safetensors +3 -0
  6. model-00004-of-00005.safetensors +3 -0
  7. model-00005-of-00005.safetensors +3 -0
  8. model.safetensors.index.json +3 -0
  9. preprocessor_config.json +21 -0
  10. tokenizer.json +3 -0
  11. tokenizer_config.json +305 -0
.gitattributes CHANGED
@@ -33,3 +33,5 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
 
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ model.safetensors.index.json filter=lfs diff=lfs merge=lfs -text
37
+ tokenizer.json filter=lfs diff=lfs merge=lfs -text
config.json ADDED
@@ -0,0 +1,718 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "Qwen3_5MoeForConditionalGeneration"
4
+ ],
5
+ "dtype": "float16",
6
+ "image_token_id": 248056,
7
+ "model_type": "qwen3_5_moe",
8
+ "quantization_config": {
9
+ "config_groups": {
10
+ "group_0": {
11
+ "format": "pack-quantized",
12
+ "input_activations": null,
13
+ "output_activations": null,
14
+ "targets": [
15
+ "Linear"
16
+ ],
17
+ "weights": {
18
+ "actorder": null,
19
+ "block_structure": null,
20
+ "dynamic": false,
21
+ "group_size": 32,
22
+ "num_bits": 4,
23
+ "observer": "mse",
24
+ "observer_kwargs": {},
25
+ "scale_dtype": null,
26
+ "strategy": "group",
27
+ "symmetric": false,
28
+ "type": "int",
29
+ "zp_dtype": "torch.int8"
30
+ }
31
+ }
32
+ },
33
+ "format": "pack-quantized",
34
+ "global_compression_ratio": null,
35
+ "ignore": [
36
+ "model.visual.blocks.0.attn.qkv",
37
+ "model.visual.blocks.0.attn.proj",
38
+ "model.visual.blocks.0.mlp.linear_fc1",
39
+ "model.visual.blocks.0.mlp.linear_fc2",
40
+ "model.visual.blocks.1.attn.qkv",
41
+ "model.visual.blocks.1.attn.proj",
42
+ "model.visual.blocks.1.mlp.linear_fc1",
43
+ "model.visual.blocks.1.mlp.linear_fc2",
44
+ "model.visual.blocks.2.attn.qkv",
45
+ "model.visual.blocks.2.attn.proj",
46
+ "model.visual.blocks.2.mlp.linear_fc1",
47
+ "model.visual.blocks.2.mlp.linear_fc2",
48
+ "model.visual.blocks.3.attn.qkv",
49
+ "model.visual.blocks.3.attn.proj",
50
+ "model.visual.blocks.3.mlp.linear_fc1",
51
+ "model.visual.blocks.3.mlp.linear_fc2",
52
+ "model.visual.blocks.4.attn.qkv",
53
+ "model.visual.blocks.4.attn.proj",
54
+ "model.visual.blocks.4.mlp.linear_fc1",
55
+ "model.visual.blocks.4.mlp.linear_fc2",
56
+ "model.visual.blocks.5.attn.qkv",
57
+ "model.visual.blocks.5.attn.proj",
58
+ "model.visual.blocks.5.mlp.linear_fc1",
59
+ "model.visual.blocks.5.mlp.linear_fc2",
60
+ "model.visual.blocks.6.attn.qkv",
61
+ "model.visual.blocks.6.attn.proj",
62
+ "model.visual.blocks.6.mlp.linear_fc1",
63
+ "model.visual.blocks.6.mlp.linear_fc2",
64
+ "model.visual.blocks.7.attn.qkv",
65
+ "model.visual.blocks.7.attn.proj",
66
+ "model.visual.blocks.7.mlp.linear_fc1",
67
+ "model.visual.blocks.7.mlp.linear_fc2",
68
+ "model.visual.blocks.8.attn.qkv",
69
+ "model.visual.blocks.8.attn.proj",
70
+ "model.visual.blocks.8.mlp.linear_fc1",
71
+ "model.visual.blocks.8.mlp.linear_fc2",
72
+ "model.visual.blocks.9.attn.qkv",
73
+ "model.visual.blocks.9.attn.proj",
74
+ "model.visual.blocks.9.mlp.linear_fc1",
75
+ "model.visual.blocks.9.mlp.linear_fc2",
76
+ "model.visual.blocks.10.attn.qkv",
77
+ "model.visual.blocks.10.attn.proj",
78
+ "model.visual.blocks.10.mlp.linear_fc1",
79
+ "model.visual.blocks.10.mlp.linear_fc2",
80
+ "model.visual.blocks.11.attn.qkv",
81
+ "model.visual.blocks.11.attn.proj",
82
+ "model.visual.blocks.11.mlp.linear_fc1",
83
+ "model.visual.blocks.11.mlp.linear_fc2",
84
+ "model.visual.blocks.12.attn.qkv",
85
+ "model.visual.blocks.12.attn.proj",
86
+ "model.visual.blocks.12.mlp.linear_fc1",
87
+ "model.visual.blocks.12.mlp.linear_fc2",
88
+ "model.visual.blocks.13.attn.qkv",
89
+ "model.visual.blocks.13.attn.proj",
90
+ "model.visual.blocks.13.mlp.linear_fc1",
91
+ "model.visual.blocks.13.mlp.linear_fc2",
92
+ "model.visual.blocks.14.attn.qkv",
93
+ "model.visual.blocks.14.attn.proj",
94
+ "model.visual.blocks.14.mlp.linear_fc1",
95
+ "model.visual.blocks.14.mlp.linear_fc2",
96
+ "model.visual.blocks.15.attn.qkv",
97
+ "model.visual.blocks.15.attn.proj",
98
+ "model.visual.blocks.15.mlp.linear_fc1",
99
+ "model.visual.blocks.15.mlp.linear_fc2",
100
+ "model.visual.blocks.16.attn.qkv",
101
+ "model.visual.blocks.16.attn.proj",
102
+ "model.visual.blocks.16.mlp.linear_fc1",
103
+ "model.visual.blocks.16.mlp.linear_fc2",
104
+ "model.visual.blocks.17.attn.qkv",
105
+ "model.visual.blocks.17.attn.proj",
106
+ "model.visual.blocks.17.mlp.linear_fc1",
107
+ "model.visual.blocks.17.mlp.linear_fc2",
108
+ "model.visual.blocks.18.attn.qkv",
109
+ "model.visual.blocks.18.attn.proj",
110
+ "model.visual.blocks.18.mlp.linear_fc1",
111
+ "model.visual.blocks.18.mlp.linear_fc2",
112
+ "model.visual.blocks.19.attn.qkv",
113
+ "model.visual.blocks.19.attn.proj",
114
+ "model.visual.blocks.19.mlp.linear_fc1",
115
+ "model.visual.blocks.19.mlp.linear_fc2",
116
+ "model.visual.blocks.20.attn.qkv",
117
+ "model.visual.blocks.20.attn.proj",
118
+ "model.visual.blocks.20.mlp.linear_fc1",
119
+ "model.visual.blocks.20.mlp.linear_fc2",
120
+ "model.visual.blocks.21.attn.qkv",
121
+ "model.visual.blocks.21.attn.proj",
122
+ "model.visual.blocks.21.mlp.linear_fc1",
123
+ "model.visual.blocks.21.mlp.linear_fc2",
124
+ "model.visual.blocks.22.attn.qkv",
125
+ "model.visual.blocks.22.attn.proj",
126
+ "model.visual.blocks.22.mlp.linear_fc1",
127
+ "model.visual.blocks.22.mlp.linear_fc2",
128
+ "model.visual.blocks.23.attn.qkv",
129
+ "model.visual.blocks.23.attn.proj",
130
+ "model.visual.blocks.23.mlp.linear_fc1",
131
+ "model.visual.blocks.23.mlp.linear_fc2",
132
+ "model.visual.blocks.24.attn.qkv",
133
+ "model.visual.blocks.24.attn.proj",
134
+ "model.visual.blocks.24.mlp.linear_fc1",
135
+ "model.visual.blocks.24.mlp.linear_fc2",
136
+ "model.visual.blocks.25.attn.qkv",
137
+ "model.visual.blocks.25.attn.proj",
138
+ "model.visual.blocks.25.mlp.linear_fc1",
139
+ "model.visual.blocks.25.mlp.linear_fc2",
140
+ "model.visual.blocks.26.attn.qkv",
141
+ "model.visual.blocks.26.attn.proj",
142
+ "model.visual.blocks.26.mlp.linear_fc1",
143
+ "model.visual.blocks.26.mlp.linear_fc2",
144
+ "model.visual.merger.linear_fc1",
145
+ "model.visual.merger.linear_fc2",
146
+ "model.language_model.layers.0.linear_attn",
147
+ "model.language_model.layers.0.linear_attn.norm",
148
+ "model.language_model.layers.0.linear_attn.out_proj",
149
+ "model.language_model.layers.0.linear_attn.in_proj_qkv",
150
+ "model.language_model.layers.0.linear_attn.in_proj_z",
151
+ "model.language_model.layers.0.linear_attn.in_proj_b",
152
+ "model.language_model.layers.0.linear_attn.in_proj_a",
153
+ "model.language_model.layers.0.mlp.gate",
154
+ "model.language_model.layers.0.mlp.shared_expert.gate_proj",
155
+ "model.language_model.layers.0.mlp.shared_expert.up_proj",
156
+ "model.language_model.layers.0.mlp.shared_expert.down_proj",
157
+ "model.language_model.layers.0.mlp.shared_expert_gate",
158
+ "model.language_model.layers.1.linear_attn",
159
+ "model.language_model.layers.1.linear_attn.norm",
160
+ "model.language_model.layers.1.linear_attn.out_proj",
161
+ "model.language_model.layers.1.linear_attn.in_proj_qkv",
162
+ "model.language_model.layers.1.linear_attn.in_proj_z",
163
+ "model.language_model.layers.1.linear_attn.in_proj_b",
164
+ "model.language_model.layers.1.linear_attn.in_proj_a",
165
+ "model.language_model.layers.1.mlp.gate",
166
+ "model.language_model.layers.1.mlp.shared_expert.gate_proj",
167
+ "model.language_model.layers.1.mlp.shared_expert.up_proj",
168
+ "model.language_model.layers.1.mlp.shared_expert.down_proj",
169
+ "model.language_model.layers.1.mlp.shared_expert_gate",
170
+ "model.language_model.layers.2.linear_attn",
171
+ "model.language_model.layers.2.linear_attn.norm",
172
+ "model.language_model.layers.2.linear_attn.out_proj",
173
+ "model.language_model.layers.2.linear_attn.in_proj_qkv",
174
+ "model.language_model.layers.2.linear_attn.in_proj_z",
175
+ "model.language_model.layers.2.linear_attn.in_proj_b",
176
+ "model.language_model.layers.2.linear_attn.in_proj_a",
177
+ "model.language_model.layers.2.mlp.gate",
178
+ "model.language_model.layers.2.mlp.shared_expert.gate_proj",
179
+ "model.language_model.layers.2.mlp.shared_expert.up_proj",
180
+ "model.language_model.layers.2.mlp.shared_expert.down_proj",
181
+ "model.language_model.layers.2.mlp.shared_expert_gate",
182
+ "model.language_model.layers.3.self_attn.q_proj",
183
+ "model.language_model.layers.3.self_attn.k_proj",
184
+ "model.language_model.layers.3.self_attn.v_proj",
185
+ "model.language_model.layers.3.self_attn.o_proj",
186
+ "model.language_model.layers.3.mlp.gate",
187
+ "model.language_model.layers.3.mlp.shared_expert.gate_proj",
188
+ "model.language_model.layers.3.mlp.shared_expert.up_proj",
189
+ "model.language_model.layers.3.mlp.shared_expert.down_proj",
190
+ "model.language_model.layers.3.mlp.shared_expert_gate",
191
+ "model.language_model.layers.4.linear_attn",
192
+ "model.language_model.layers.4.linear_attn.norm",
193
+ "model.language_model.layers.4.linear_attn.out_proj",
194
+ "model.language_model.layers.4.linear_attn.in_proj_qkv",
195
+ "model.language_model.layers.4.linear_attn.in_proj_z",
196
+ "model.language_model.layers.4.linear_attn.in_proj_b",
197
+ "model.language_model.layers.4.linear_attn.in_proj_a",
198
+ "model.language_model.layers.4.mlp.gate",
199
+ "model.language_model.layers.4.mlp.shared_expert.gate_proj",
200
+ "model.language_model.layers.4.mlp.shared_expert.up_proj",
201
+ "model.language_model.layers.4.mlp.shared_expert.down_proj",
202
+ "model.language_model.layers.4.mlp.shared_expert_gate",
203
+ "model.language_model.layers.5.linear_attn",
204
+ "model.language_model.layers.5.linear_attn.norm",
205
+ "model.language_model.layers.5.linear_attn.out_proj",
206
+ "model.language_model.layers.5.linear_attn.in_proj_qkv",
207
+ "model.language_model.layers.5.linear_attn.in_proj_z",
208
+ "model.language_model.layers.5.linear_attn.in_proj_b",
209
+ "model.language_model.layers.5.linear_attn.in_proj_a",
210
+ "model.language_model.layers.5.mlp.gate",
211
+ "model.language_model.layers.5.mlp.shared_expert.gate_proj",
212
+ "model.language_model.layers.5.mlp.shared_expert.up_proj",
213
+ "model.language_model.layers.5.mlp.shared_expert.down_proj",
214
+ "model.language_model.layers.5.mlp.shared_expert_gate",
215
+ "model.language_model.layers.6.linear_attn",
216
+ "model.language_model.layers.6.linear_attn.norm",
217
+ "model.language_model.layers.6.linear_attn.out_proj",
218
+ "model.language_model.layers.6.linear_attn.in_proj_qkv",
219
+ "model.language_model.layers.6.linear_attn.in_proj_z",
220
+ "model.language_model.layers.6.linear_attn.in_proj_b",
221
+ "model.language_model.layers.6.linear_attn.in_proj_a",
222
+ "model.language_model.layers.6.mlp.gate",
223
+ "model.language_model.layers.6.mlp.shared_expert.gate_proj",
224
+ "model.language_model.layers.6.mlp.shared_expert.up_proj",
225
+ "model.language_model.layers.6.mlp.shared_expert.down_proj",
226
+ "model.language_model.layers.6.mlp.shared_expert_gate",
227
+ "model.language_model.layers.7.self_attn.q_proj",
228
+ "model.language_model.layers.7.self_attn.k_proj",
229
+ "model.language_model.layers.7.self_attn.v_proj",
230
+ "model.language_model.layers.7.self_attn.o_proj",
231
+ "model.language_model.layers.7.mlp.gate",
232
+ "model.language_model.layers.7.mlp.shared_expert.gate_proj",
233
+ "model.language_model.layers.7.mlp.shared_expert.up_proj",
234
+ "model.language_model.layers.7.mlp.shared_expert.down_proj",
235
+ "model.language_model.layers.7.mlp.shared_expert_gate",
236
+ "model.language_model.layers.8.linear_attn",
237
+ "model.language_model.layers.8.linear_attn.norm",
238
+ "model.language_model.layers.8.linear_attn.out_proj",
239
+ "model.language_model.layers.8.linear_attn.in_proj_qkv",
240
+ "model.language_model.layers.8.linear_attn.in_proj_z",
241
+ "model.language_model.layers.8.linear_attn.in_proj_b",
242
+ "model.language_model.layers.8.linear_attn.in_proj_a",
243
+ "model.language_model.layers.8.mlp.gate",
244
+ "model.language_model.layers.8.mlp.shared_expert.gate_proj",
245
+ "model.language_model.layers.8.mlp.shared_expert.up_proj",
246
+ "model.language_model.layers.8.mlp.shared_expert.down_proj",
247
+ "model.language_model.layers.8.mlp.shared_expert_gate",
248
+ "model.language_model.layers.9.linear_attn",
249
+ "model.language_model.layers.9.linear_attn.norm",
250
+ "model.language_model.layers.9.linear_attn.out_proj",
251
+ "model.language_model.layers.9.linear_attn.in_proj_qkv",
252
+ "model.language_model.layers.9.linear_attn.in_proj_z",
253
+ "model.language_model.layers.9.linear_attn.in_proj_b",
254
+ "model.language_model.layers.9.linear_attn.in_proj_a",
255
+ "model.language_model.layers.9.mlp.gate",
256
+ "model.language_model.layers.9.mlp.shared_expert.gate_proj",
257
+ "model.language_model.layers.9.mlp.shared_expert.up_proj",
258
+ "model.language_model.layers.9.mlp.shared_expert.down_proj",
259
+ "model.language_model.layers.9.mlp.shared_expert_gate",
260
+ "model.language_model.layers.10.linear_attn",
261
+ "model.language_model.layers.10.linear_attn.norm",
262
+ "model.language_model.layers.10.linear_attn.out_proj",
263
+ "model.language_model.layers.10.linear_attn.in_proj_qkv",
264
+ "model.language_model.layers.10.linear_attn.in_proj_z",
265
+ "model.language_model.layers.10.linear_attn.in_proj_b",
266
+ "model.language_model.layers.10.linear_attn.in_proj_a",
267
+ "model.language_model.layers.10.mlp.gate",
268
+ "model.language_model.layers.10.mlp.shared_expert.gate_proj",
269
+ "model.language_model.layers.10.mlp.shared_expert.up_proj",
270
+ "model.language_model.layers.10.mlp.shared_expert.down_proj",
271
+ "model.language_model.layers.10.mlp.shared_expert_gate",
272
+ "model.language_model.layers.11.self_attn.q_proj",
273
+ "model.language_model.layers.11.self_attn.k_proj",
274
+ "model.language_model.layers.11.self_attn.v_proj",
275
+ "model.language_model.layers.11.self_attn.o_proj",
276
+ "model.language_model.layers.11.mlp.gate",
277
+ "model.language_model.layers.11.mlp.shared_expert.gate_proj",
278
+ "model.language_model.layers.11.mlp.shared_expert.up_proj",
279
+ "model.language_model.layers.11.mlp.shared_expert.down_proj",
280
+ "model.language_model.layers.11.mlp.shared_expert_gate",
281
+ "model.language_model.layers.12.linear_attn",
282
+ "model.language_model.layers.12.linear_attn.norm",
283
+ "model.language_model.layers.12.linear_attn.out_proj",
284
+ "model.language_model.layers.12.linear_attn.in_proj_qkv",
285
+ "model.language_model.layers.12.linear_attn.in_proj_z",
286
+ "model.language_model.layers.12.linear_attn.in_proj_b",
287
+ "model.language_model.layers.12.linear_attn.in_proj_a",
288
+ "model.language_model.layers.12.mlp.gate",
289
+ "model.language_model.layers.12.mlp.shared_expert.gate_proj",
290
+ "model.language_model.layers.12.mlp.shared_expert.up_proj",
291
+ "model.language_model.layers.12.mlp.shared_expert.down_proj",
292
+ "model.language_model.layers.12.mlp.shared_expert_gate",
293
+ "model.language_model.layers.13.linear_attn",
294
+ "model.language_model.layers.13.linear_attn.norm",
295
+ "model.language_model.layers.13.linear_attn.out_proj",
296
+ "model.language_model.layers.13.linear_attn.in_proj_qkv",
297
+ "model.language_model.layers.13.linear_attn.in_proj_z",
298
+ "model.language_model.layers.13.linear_attn.in_proj_b",
299
+ "model.language_model.layers.13.linear_attn.in_proj_a",
300
+ "model.language_model.layers.13.mlp.gate",
301
+ "model.language_model.layers.13.mlp.shared_expert.gate_proj",
302
+ "model.language_model.layers.13.mlp.shared_expert.up_proj",
303
+ "model.language_model.layers.13.mlp.shared_expert.down_proj",
304
+ "model.language_model.layers.13.mlp.shared_expert_gate",
305
+ "model.language_model.layers.14.linear_attn",
306
+ "model.language_model.layers.14.linear_attn.norm",
307
+ "model.language_model.layers.14.linear_attn.out_proj",
308
+ "model.language_model.layers.14.linear_attn.in_proj_qkv",
309
+ "model.language_model.layers.14.linear_attn.in_proj_z",
310
+ "model.language_model.layers.14.linear_attn.in_proj_b",
311
+ "model.language_model.layers.14.linear_attn.in_proj_a",
312
+ "model.language_model.layers.14.mlp.gate",
313
+ "model.language_model.layers.14.mlp.shared_expert.gate_proj",
314
+ "model.language_model.layers.14.mlp.shared_expert.up_proj",
315
+ "model.language_model.layers.14.mlp.shared_expert.down_proj",
316
+ "model.language_model.layers.14.mlp.shared_expert_gate",
317
+ "model.language_model.layers.15.self_attn.q_proj",
318
+ "model.language_model.layers.15.self_attn.k_proj",
319
+ "model.language_model.layers.15.self_attn.v_proj",
320
+ "model.language_model.layers.15.self_attn.o_proj",
321
+ "model.language_model.layers.15.mlp.gate",
322
+ "model.language_model.layers.15.mlp.shared_expert.gate_proj",
323
+ "model.language_model.layers.15.mlp.shared_expert.up_proj",
324
+ "model.language_model.layers.15.mlp.shared_expert.down_proj",
325
+ "model.language_model.layers.15.mlp.shared_expert_gate",
326
+ "model.language_model.layers.16.linear_attn",
327
+ "model.language_model.layers.16.linear_attn.norm",
328
+ "model.language_model.layers.16.linear_attn.out_proj",
329
+ "model.language_model.layers.16.linear_attn.in_proj_qkv",
330
+ "model.language_model.layers.16.linear_attn.in_proj_z",
331
+ "model.language_model.layers.16.linear_attn.in_proj_b",
332
+ "model.language_model.layers.16.linear_attn.in_proj_a",
333
+ "model.language_model.layers.16.mlp.gate",
334
+ "model.language_model.layers.16.mlp.shared_expert.gate_proj",
335
+ "model.language_model.layers.16.mlp.shared_expert.up_proj",
336
+ "model.language_model.layers.16.mlp.shared_expert.down_proj",
337
+ "model.language_model.layers.16.mlp.shared_expert_gate",
338
+ "model.language_model.layers.17.linear_attn",
339
+ "model.language_model.layers.17.linear_attn.norm",
340
+ "model.language_model.layers.17.linear_attn.out_proj",
341
+ "model.language_model.layers.17.linear_attn.in_proj_qkv",
342
+ "model.language_model.layers.17.linear_attn.in_proj_z",
343
+ "model.language_model.layers.17.linear_attn.in_proj_b",
344
+ "model.language_model.layers.17.linear_attn.in_proj_a",
345
+ "model.language_model.layers.17.mlp.gate",
346
+ "model.language_model.layers.17.mlp.shared_expert.gate_proj",
347
+ "model.language_model.layers.17.mlp.shared_expert.up_proj",
348
+ "model.language_model.layers.17.mlp.shared_expert.down_proj",
349
+ "model.language_model.layers.17.mlp.shared_expert_gate",
350
+ "model.language_model.layers.18.linear_attn",
351
+ "model.language_model.layers.18.linear_attn.norm",
352
+ "model.language_model.layers.18.linear_attn.out_proj",
353
+ "model.language_model.layers.18.linear_attn.in_proj_qkv",
354
+ "model.language_model.layers.18.linear_attn.in_proj_z",
355
+ "model.language_model.layers.18.linear_attn.in_proj_b",
356
+ "model.language_model.layers.18.linear_attn.in_proj_a",
357
+ "model.language_model.layers.18.mlp.gate",
358
+ "model.language_model.layers.18.mlp.shared_expert.gate_proj",
359
+ "model.language_model.layers.18.mlp.shared_expert.up_proj",
360
+ "model.language_model.layers.18.mlp.shared_expert.down_proj",
361
+ "model.language_model.layers.18.mlp.shared_expert_gate",
362
+ "model.language_model.layers.19.self_attn.q_proj",
363
+ "model.language_model.layers.19.self_attn.k_proj",
364
+ "model.language_model.layers.19.self_attn.v_proj",
365
+ "model.language_model.layers.19.self_attn.o_proj",
366
+ "model.language_model.layers.19.mlp.gate",
367
+ "model.language_model.layers.19.mlp.shared_expert.gate_proj",
368
+ "model.language_model.layers.19.mlp.shared_expert.up_proj",
369
+ "model.language_model.layers.19.mlp.shared_expert.down_proj",
370
+ "model.language_model.layers.19.mlp.shared_expert_gate",
371
+ "model.language_model.layers.20.linear_attn",
372
+ "model.language_model.layers.20.linear_attn.norm",
373
+ "model.language_model.layers.20.linear_attn.out_proj",
374
+ "model.language_model.layers.20.linear_attn.in_proj_qkv",
375
+ "model.language_model.layers.20.linear_attn.in_proj_z",
376
+ "model.language_model.layers.20.linear_attn.in_proj_b",
377
+ "model.language_model.layers.20.linear_attn.in_proj_a",
378
+ "model.language_model.layers.20.mlp.gate",
379
+ "model.language_model.layers.20.mlp.shared_expert.gate_proj",
380
+ "model.language_model.layers.20.mlp.shared_expert.up_proj",
381
+ "model.language_model.layers.20.mlp.shared_expert.down_proj",
382
+ "model.language_model.layers.20.mlp.shared_expert_gate",
383
+ "model.language_model.layers.21.linear_attn",
384
+ "model.language_model.layers.21.linear_attn.norm",
385
+ "model.language_model.layers.21.linear_attn.out_proj",
386
+ "model.language_model.layers.21.linear_attn.in_proj_qkv",
387
+ "model.language_model.layers.21.linear_attn.in_proj_z",
388
+ "model.language_model.layers.21.linear_attn.in_proj_b",
389
+ "model.language_model.layers.21.linear_attn.in_proj_a",
390
+ "model.language_model.layers.21.mlp.gate",
391
+ "model.language_model.layers.21.mlp.shared_expert.gate_proj",
392
+ "model.language_model.layers.21.mlp.shared_expert.up_proj",
393
+ "model.language_model.layers.21.mlp.shared_expert.down_proj",
394
+ "model.language_model.layers.21.mlp.shared_expert_gate",
395
+ "model.language_model.layers.22.linear_attn",
396
+ "model.language_model.layers.22.linear_attn.norm",
397
+ "model.language_model.layers.22.linear_attn.out_proj",
398
+ "model.language_model.layers.22.linear_attn.in_proj_qkv",
399
+ "model.language_model.layers.22.linear_attn.in_proj_z",
400
+ "model.language_model.layers.22.linear_attn.in_proj_b",
401
+ "model.language_model.layers.22.linear_attn.in_proj_a",
402
+ "model.language_model.layers.22.mlp.gate",
403
+ "model.language_model.layers.22.mlp.shared_expert.gate_proj",
404
+ "model.language_model.layers.22.mlp.shared_expert.up_proj",
405
+ "model.language_model.layers.22.mlp.shared_expert.down_proj",
406
+ "model.language_model.layers.22.mlp.shared_expert_gate",
407
+ "model.language_model.layers.23.self_attn.q_proj",
408
+ "model.language_model.layers.23.self_attn.k_proj",
409
+ "model.language_model.layers.23.self_attn.v_proj",
410
+ "model.language_model.layers.23.self_attn.o_proj",
411
+ "model.language_model.layers.23.mlp.gate",
412
+ "model.language_model.layers.23.mlp.shared_expert.gate_proj",
413
+ "model.language_model.layers.23.mlp.shared_expert.up_proj",
414
+ "model.language_model.layers.23.mlp.shared_expert.down_proj",
415
+ "model.language_model.layers.23.mlp.shared_expert_gate",
416
+ "model.language_model.layers.24.linear_attn",
417
+ "model.language_model.layers.24.linear_attn.norm",
418
+ "model.language_model.layers.24.linear_attn.out_proj",
419
+ "model.language_model.layers.24.linear_attn.in_proj_qkv",
420
+ "model.language_model.layers.24.linear_attn.in_proj_z",
421
+ "model.language_model.layers.24.linear_attn.in_proj_b",
422
+ "model.language_model.layers.24.linear_attn.in_proj_a",
423
+ "model.language_model.layers.24.mlp.gate",
424
+ "model.language_model.layers.24.mlp.shared_expert.gate_proj",
425
+ "model.language_model.layers.24.mlp.shared_expert.up_proj",
426
+ "model.language_model.layers.24.mlp.shared_expert.down_proj",
427
+ "model.language_model.layers.24.mlp.shared_expert_gate",
428
+ "model.language_model.layers.25.linear_attn",
429
+ "model.language_model.layers.25.linear_attn.norm",
430
+ "model.language_model.layers.25.linear_attn.out_proj",
431
+ "model.language_model.layers.25.linear_attn.in_proj_qkv",
432
+ "model.language_model.layers.25.linear_attn.in_proj_z",
433
+ "model.language_model.layers.25.linear_attn.in_proj_b",
434
+ "model.language_model.layers.25.linear_attn.in_proj_a",
435
+ "model.language_model.layers.25.mlp.gate",
436
+ "model.language_model.layers.25.mlp.shared_expert.gate_proj",
437
+ "model.language_model.layers.25.mlp.shared_expert.up_proj",
438
+ "model.language_model.layers.25.mlp.shared_expert.down_proj",
439
+ "model.language_model.layers.25.mlp.shared_expert_gate",
440
+ "model.language_model.layers.26.linear_attn",
441
+ "model.language_model.layers.26.linear_attn.norm",
442
+ "model.language_model.layers.26.linear_attn.out_proj",
443
+ "model.language_model.layers.26.linear_attn.in_proj_qkv",
444
+ "model.language_model.layers.26.linear_attn.in_proj_z",
445
+ "model.language_model.layers.26.linear_attn.in_proj_b",
446
+ "model.language_model.layers.26.linear_attn.in_proj_a",
447
+ "model.language_model.layers.26.mlp.gate",
448
+ "model.language_model.layers.26.mlp.shared_expert.gate_proj",
449
+ "model.language_model.layers.26.mlp.shared_expert.up_proj",
450
+ "model.language_model.layers.26.mlp.shared_expert.down_proj",
451
+ "model.language_model.layers.26.mlp.shared_expert_gate",
452
+ "model.language_model.layers.27.self_attn.q_proj",
453
+ "model.language_model.layers.27.self_attn.k_proj",
454
+ "model.language_model.layers.27.self_attn.v_proj",
455
+ "model.language_model.layers.27.self_attn.o_proj",
456
+ "model.language_model.layers.27.mlp.gate",
457
+ "model.language_model.layers.27.mlp.shared_expert.gate_proj",
458
+ "model.language_model.layers.27.mlp.shared_expert.up_proj",
459
+ "model.language_model.layers.27.mlp.shared_expert.down_proj",
460
+ "model.language_model.layers.27.mlp.shared_expert_gate",
461
+ "model.language_model.layers.28.linear_attn",
462
+ "model.language_model.layers.28.linear_attn.norm",
463
+ "model.language_model.layers.28.linear_attn.out_proj",
464
+ "model.language_model.layers.28.linear_attn.in_proj_qkv",
465
+ "model.language_model.layers.28.linear_attn.in_proj_z",
466
+ "model.language_model.layers.28.linear_attn.in_proj_b",
467
+ "model.language_model.layers.28.linear_attn.in_proj_a",
468
+ "model.language_model.layers.28.mlp.gate",
469
+ "model.language_model.layers.28.mlp.shared_expert.gate_proj",
470
+ "model.language_model.layers.28.mlp.shared_expert.up_proj",
471
+ "model.language_model.layers.28.mlp.shared_expert.down_proj",
472
+ "model.language_model.layers.28.mlp.shared_expert_gate",
473
+ "model.language_model.layers.29.linear_attn",
474
+ "model.language_model.layers.29.linear_attn.norm",
475
+ "model.language_model.layers.29.linear_attn.out_proj",
476
+ "model.language_model.layers.29.linear_attn.in_proj_qkv",
477
+ "model.language_model.layers.29.linear_attn.in_proj_z",
478
+ "model.language_model.layers.29.linear_attn.in_proj_b",
479
+ "model.language_model.layers.29.linear_attn.in_proj_a",
480
+ "model.language_model.layers.29.mlp.gate",
481
+ "model.language_model.layers.29.mlp.shared_expert.gate_proj",
482
+ "model.language_model.layers.29.mlp.shared_expert.up_proj",
483
+ "model.language_model.layers.29.mlp.shared_expert.down_proj",
484
+ "model.language_model.layers.29.mlp.shared_expert_gate",
485
+ "model.language_model.layers.30.linear_attn",
486
+ "model.language_model.layers.30.linear_attn.norm",
487
+ "model.language_model.layers.30.linear_attn.out_proj",
488
+ "model.language_model.layers.30.linear_attn.in_proj_qkv",
489
+ "model.language_model.layers.30.linear_attn.in_proj_z",
490
+ "model.language_model.layers.30.linear_attn.in_proj_b",
491
+ "model.language_model.layers.30.linear_attn.in_proj_a",
492
+ "model.language_model.layers.30.mlp.gate",
493
+ "model.language_model.layers.30.mlp.shared_expert.gate_proj",
494
+ "model.language_model.layers.30.mlp.shared_expert.up_proj",
495
+ "model.language_model.layers.30.mlp.shared_expert.down_proj",
496
+ "model.language_model.layers.30.mlp.shared_expert_gate",
497
+ "model.language_model.layers.31.self_attn.q_proj",
498
+ "model.language_model.layers.31.self_attn.k_proj",
499
+ "model.language_model.layers.31.self_attn.v_proj",
500
+ "model.language_model.layers.31.self_attn.o_proj",
501
+ "model.language_model.layers.31.mlp.gate",
502
+ "model.language_model.layers.31.mlp.shared_expert.gate_proj",
503
+ "model.language_model.layers.31.mlp.shared_expert.up_proj",
504
+ "model.language_model.layers.31.mlp.shared_expert.down_proj",
505
+ "model.language_model.layers.31.mlp.shared_expert_gate",
506
+ "model.language_model.layers.32.linear_attn",
507
+ "model.language_model.layers.32.linear_attn.norm",
508
+ "model.language_model.layers.32.linear_attn.out_proj",
509
+ "model.language_model.layers.32.linear_attn.in_proj_qkv",
510
+ "model.language_model.layers.32.linear_attn.in_proj_z",
511
+ "model.language_model.layers.32.linear_attn.in_proj_b",
512
+ "model.language_model.layers.32.linear_attn.in_proj_a",
513
+ "model.language_model.layers.32.mlp.gate",
514
+ "model.language_model.layers.32.mlp.shared_expert.gate_proj",
515
+ "model.language_model.layers.32.mlp.shared_expert.up_proj",
516
+ "model.language_model.layers.32.mlp.shared_expert.down_proj",
517
+ "model.language_model.layers.32.mlp.shared_expert_gate",
518
+ "model.language_model.layers.33.linear_attn",
519
+ "model.language_model.layers.33.linear_attn.norm",
520
+ "model.language_model.layers.33.linear_attn.out_proj",
521
+ "model.language_model.layers.33.linear_attn.in_proj_qkv",
522
+ "model.language_model.layers.33.linear_attn.in_proj_z",
523
+ "model.language_model.layers.33.linear_attn.in_proj_b",
524
+ "model.language_model.layers.33.linear_attn.in_proj_a",
525
+ "model.language_model.layers.33.mlp.gate",
526
+ "model.language_model.layers.33.mlp.shared_expert.gate_proj",
527
+ "model.language_model.layers.33.mlp.shared_expert.up_proj",
528
+ "model.language_model.layers.33.mlp.shared_expert.down_proj",
529
+ "model.language_model.layers.33.mlp.shared_expert_gate",
530
+ "model.language_model.layers.34.linear_attn",
531
+ "model.language_model.layers.34.linear_attn.norm",
532
+ "model.language_model.layers.34.linear_attn.out_proj",
533
+ "model.language_model.layers.34.linear_attn.in_proj_qkv",
534
+ "model.language_model.layers.34.linear_attn.in_proj_z",
535
+ "model.language_model.layers.34.linear_attn.in_proj_b",
536
+ "model.language_model.layers.34.linear_attn.in_proj_a",
537
+ "model.language_model.layers.34.mlp.gate",
538
+ "model.language_model.layers.34.mlp.shared_expert.gate_proj",
539
+ "model.language_model.layers.34.mlp.shared_expert.up_proj",
540
+ "model.language_model.layers.34.mlp.shared_expert.down_proj",
541
+ "model.language_model.layers.34.mlp.shared_expert_gate",
542
+ "model.language_model.layers.35.self_attn.q_proj",
543
+ "model.language_model.layers.35.self_attn.k_proj",
544
+ "model.language_model.layers.35.self_attn.v_proj",
545
+ "model.language_model.layers.35.self_attn.o_proj",
546
+ "model.language_model.layers.35.mlp.gate",
547
+ "model.language_model.layers.35.mlp.shared_expert.gate_proj",
548
+ "model.language_model.layers.35.mlp.shared_expert.up_proj",
549
+ "model.language_model.layers.35.mlp.shared_expert.down_proj",
550
+ "model.language_model.layers.35.mlp.shared_expert_gate",
551
+ "model.language_model.layers.36.linear_attn",
552
+ "model.language_model.layers.36.linear_attn.norm",
553
+ "model.language_model.layers.36.linear_attn.out_proj",
554
+ "model.language_model.layers.36.linear_attn.in_proj_qkv",
555
+ "model.language_model.layers.36.linear_attn.in_proj_z",
556
+ "model.language_model.layers.36.linear_attn.in_proj_b",
557
+ "model.language_model.layers.36.linear_attn.in_proj_a",
558
+ "model.language_model.layers.36.mlp.gate",
559
+ "model.language_model.layers.36.mlp.shared_expert.gate_proj",
560
+ "model.language_model.layers.36.mlp.shared_expert.up_proj",
561
+ "model.language_model.layers.36.mlp.shared_expert.down_proj",
562
+ "model.language_model.layers.36.mlp.shared_expert_gate",
563
+ "model.language_model.layers.37.linear_attn",
564
+ "model.language_model.layers.37.linear_attn.norm",
565
+ "model.language_model.layers.37.linear_attn.out_proj",
566
+ "model.language_model.layers.37.linear_attn.in_proj_qkv",
567
+ "model.language_model.layers.37.linear_attn.in_proj_z",
568
+ "model.language_model.layers.37.linear_attn.in_proj_b",
569
+ "model.language_model.layers.37.linear_attn.in_proj_a",
570
+ "model.language_model.layers.37.mlp.gate",
571
+ "model.language_model.layers.37.mlp.shared_expert.gate_proj",
572
+ "model.language_model.layers.37.mlp.shared_expert.up_proj",
573
+ "model.language_model.layers.37.mlp.shared_expert.down_proj",
574
+ "model.language_model.layers.37.mlp.shared_expert_gate",
575
+ "model.language_model.layers.38.linear_attn",
576
+ "model.language_model.layers.38.linear_attn.norm",
577
+ "model.language_model.layers.38.linear_attn.out_proj",
578
+ "model.language_model.layers.38.linear_attn.in_proj_qkv",
579
+ "model.language_model.layers.38.linear_attn.in_proj_z",
580
+ "model.language_model.layers.38.linear_attn.in_proj_b",
581
+ "model.language_model.layers.38.linear_attn.in_proj_a",
582
+ "model.language_model.layers.38.mlp.gate",
583
+ "model.language_model.layers.38.mlp.shared_expert.gate_proj",
584
+ "model.language_model.layers.38.mlp.shared_expert.up_proj",
585
+ "model.language_model.layers.38.mlp.shared_expert.down_proj",
586
+ "model.language_model.layers.38.mlp.shared_expert_gate",
587
+ "model.language_model.layers.39.self_attn.q_proj",
588
+ "model.language_model.layers.39.self_attn.k_proj",
589
+ "model.language_model.layers.39.self_attn.v_proj",
590
+ "model.language_model.layers.39.self_attn.o_proj",
591
+ "model.language_model.layers.39.mlp.gate",
592
+ "model.language_model.layers.39.mlp.shared_expert.gate_proj",
593
+ "model.language_model.layers.39.mlp.shared_expert.up_proj",
594
+ "model.language_model.layers.39.mlp.shared_expert.down_proj",
595
+ "model.language_model.layers.39.mlp.shared_expert_gate",
596
+ "lm_head"
597
+ ],
598
+ "kv_cache_scheme": null,
599
+ "quant_method": "compressed-tensors",
600
+ "quantization_status": "compressed",
601
+ "sparsity_config": {},
602
+ "transform_config": {},
603
+ "version": "0.17.2.dev26+g1829ea9"
604
+ },
605
+ "text_config": {
606
+ "attention_bias": false,
607
+ "attention_dropout": 0.0,
608
+ "attn_output_gate": true,
609
+ "bos_token_id": 248044,
610
+ "dtype": "float16",
611
+ "eos_token_id": 248044,
612
+ "full_attention_interval": 4,
613
+ "head_dim": 256,
614
+ "hidden_act": "silu",
615
+ "hidden_size": 2048,
616
+ "initializer_range": 0.02,
617
+ "layer_types": [
618
+ "linear_attention",
619
+ "linear_attention",
620
+ "linear_attention",
621
+ "full_attention",
622
+ "linear_attention",
623
+ "linear_attention",
624
+ "linear_attention",
625
+ "full_attention",
626
+ "linear_attention",
627
+ "linear_attention",
628
+ "linear_attention",
629
+ "full_attention",
630
+ "linear_attention",
631
+ "linear_attention",
632
+ "linear_attention",
633
+ "full_attention",
634
+ "linear_attention",
635
+ "linear_attention",
636
+ "linear_attention",
637
+ "full_attention",
638
+ "linear_attention",
639
+ "linear_attention",
640
+ "linear_attention",
641
+ "full_attention",
642
+ "linear_attention",
643
+ "linear_attention",
644
+ "linear_attention",
645
+ "full_attention",
646
+ "linear_attention",
647
+ "linear_attention",
648
+ "linear_attention",
649
+ "full_attention",
650
+ "linear_attention",
651
+ "linear_attention",
652
+ "linear_attention",
653
+ "full_attention",
654
+ "linear_attention",
655
+ "linear_attention",
656
+ "linear_attention",
657
+ "full_attention"
658
+ ],
659
+ "linear_conv_kernel_dim": 4,
660
+ "linear_key_head_dim": 128,
661
+ "linear_num_key_heads": 16,
662
+ "linear_num_value_heads": 32,
663
+ "linear_value_head_dim": 128,
664
+ "mamba_ssm_dtype": "float32",
665
+ "max_position_embeddings": 262144,
666
+ "model_type": "qwen3_5_moe_text",
667
+ "moe_intermediate_size": 512,
668
+ "mtp_num_hidden_layers": 0,
669
+ "mtp_use_dedicated_embeddings": false,
670
+ "num_attention_heads": 16,
671
+ "num_experts": 256,
672
+ "num_experts_per_tok": 8,
673
+ "num_hidden_layers": 40,
674
+ "num_key_value_heads": 2,
675
+ "output_router_logits": false,
676
+ "pad_token_id": null,
677
+ "partial_rotary_factor": 0.25,
678
+ "rms_norm_eps": 1e-06,
679
+ "rope_parameters": {
680
+ "mrope_interleaved": true,
681
+ "mrope_section": [
682
+ 11,
683
+ 11,
684
+ 10
685
+ ],
686
+ "partial_rotary_factor": 0.25,
687
+ "rope_theta": 10000000,
688
+ "rope_type": "default"
689
+ },
690
+ "router_aux_loss_coef": 0.001,
691
+ "shared_expert_intermediate_size": 512,
692
+ "tie_word_embeddings": false,
693
+ "use_cache": true,
694
+ "vocab_size": 248320
695
+ },
696
+ "tie_word_embeddings": false,
697
+ "transformers_version": "5.13.0.dev0",
698
+ "video_token_id": 248057,
699
+ "vision_config": {
700
+ "deepstack_visual_indexes": [],
701
+ "depth": 27,
702
+ "dtype": "float16",
703
+ "hidden_act": "gelu_pytorch_tanh",
704
+ "hidden_size": 1152,
705
+ "in_channels": 3,
706
+ "initializer_range": 0.02,
707
+ "intermediate_size": 4304,
708
+ "model_type": "qwen3_5_moe_vision",
709
+ "num_heads": 16,
710
+ "num_position_embeddings": 2304,
711
+ "out_hidden_size": 2048,
712
+ "patch_size": 16,
713
+ "spatial_merge_size": 2,
714
+ "temporal_patch_size": 2
715
+ },
716
+ "vision_end_token_id": 248054,
717
+ "vision_start_token_id": 248053
718
+ }
model-00001-of-00005.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:84033db2debaecfce45a9ca10584b25ce62be3b45cb0def3ca0e03e069ee06dd
3
+ size 5371198744
model-00002-of-00005.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cc2de5d24855df6f1eeee13b044f3041f933377ea3eb1b01f6598dc25f6e04c5
3
+ size 5373023304
model-00003-of-00005.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f2822e4224f60beb468a691a6c5b41c0e7ab3fe016ea793d72ee2fa42f820438
3
+ size 5372856504
model-00004-of-00005.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5c670fa260a0fd139f4231dadedb6aba56591c2aaad70f3276ff40f9942c8e04
3
+ size 5373021968
model-00005-of-00005.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0ed5778cd292caa4b0855128764f2bbb6706a68ae45a2bcd2ddd238d5e600d26
3
+ size 2940307952
model.safetensors.index.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:62e4c9c780f9f453c355b07078a726a8423323cdb10890524a393823fc7bef05
3
+ size 13993420
preprocessor_config.json ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "size": {
3
+ "longest_edge": 16777216,
4
+ "shortest_edge": 65536
5
+ },
6
+ "patch_size": 16,
7
+ "temporal_patch_size": 2,
8
+ "merge_size": 2,
9
+ "image_mean": [
10
+ 0.5,
11
+ 0.5,
12
+ 0.5
13
+ ],
14
+ "image_std": [
15
+ 0.5,
16
+ 0.5,
17
+ 0.5
18
+ ],
19
+ "processor_class": "Qwen3VLProcessor",
20
+ "image_processor_type": "Qwen2VLImageProcessorFast"
21
+ }
tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5f9e4d4901a92b997e463c1f46055088b6cca5ca61a6522d1b9f64c4bb81cb42
3
+ size 12807982
tokenizer_config.json ADDED
@@ -0,0 +1,305 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "added_tokens_decoder": {
4
+ "248044": {
5
+ "content": "<|endoftext|>",
6
+ "lstrip": false,
7
+ "normalized": false,
8
+ "rstrip": false,
9
+ "single_word": false,
10
+ "special": true
11
+ },
12
+ "248045": {
13
+ "content": "<|im_start|>",
14
+ "lstrip": false,
15
+ "normalized": false,
16
+ "rstrip": false,
17
+ "single_word": false,
18
+ "special": true
19
+ },
20
+ "248046": {
21
+ "content": "<|im_end|>",
22
+ "lstrip": false,
23
+ "normalized": false,
24
+ "rstrip": false,
25
+ "single_word": false,
26
+ "special": true
27
+ },
28
+ "248047": {
29
+ "content": "<|object_ref_start|>",
30
+ "lstrip": false,
31
+ "normalized": false,
32
+ "rstrip": false,
33
+ "single_word": false,
34
+ "special": true
35
+ },
36
+ "248048": {
37
+ "content": "<|object_ref_end|>",
38
+ "lstrip": false,
39
+ "normalized": false,
40
+ "rstrip": false,
41
+ "single_word": false,
42
+ "special": true
43
+ },
44
+ "248049": {
45
+ "content": "<|box_start|>",
46
+ "lstrip": false,
47
+ "normalized": false,
48
+ "rstrip": false,
49
+ "single_word": false,
50
+ "special": true
51
+ },
52
+ "248050": {
53
+ "content": "<|box_end|>",
54
+ "lstrip": false,
55
+ "normalized": false,
56
+ "rstrip": false,
57
+ "single_word": false,
58
+ "special": true
59
+ },
60
+ "248051": {
61
+ "content": "<|quad_start|>",
62
+ "lstrip": false,
63
+ "normalized": false,
64
+ "rstrip": false,
65
+ "single_word": false,
66
+ "special": true
67
+ },
68
+ "248052": {
69
+ "content": "<|quad_end|>",
70
+ "lstrip": false,
71
+ "normalized": false,
72
+ "rstrip": false,
73
+ "single_word": false,
74
+ "special": true
75
+ },
76
+ "248053": {
77
+ "content": "<|vision_start|>",
78
+ "lstrip": false,
79
+ "normalized": false,
80
+ "rstrip": false,
81
+ "single_word": false,
82
+ "special": true
83
+ },
84
+ "248054": {
85
+ "content": "<|vision_end|>",
86
+ "lstrip": false,
87
+ "normalized": false,
88
+ "rstrip": false,
89
+ "single_word": false,
90
+ "special": true
91
+ },
92
+ "248055": {
93
+ "content": "<|vision_pad|>",
94
+ "lstrip": false,
95
+ "normalized": false,
96
+ "rstrip": false,
97
+ "single_word": false,
98
+ "special": true
99
+ },
100
+ "248056": {
101
+ "content": "<|image_pad|>",
102
+ "lstrip": false,
103
+ "normalized": false,
104
+ "rstrip": false,
105
+ "single_word": false,
106
+ "special": true
107
+ },
108
+ "248057": {
109
+ "content": "<|video_pad|>",
110
+ "lstrip": false,
111
+ "normalized": false,
112
+ "rstrip": false,
113
+ "single_word": false,
114
+ "special": true
115
+ },
116
+ "248058": {
117
+ "content": "<tool_call>",
118
+ "lstrip": false,
119
+ "normalized": false,
120
+ "rstrip": false,
121
+ "single_word": false,
122
+ "special": false
123
+ },
124
+ "248059": {
125
+ "content": "</tool_call>",
126
+ "lstrip": false,
127
+ "normalized": false,
128
+ "rstrip": false,
129
+ "single_word": false,
130
+ "special": false
131
+ },
132
+ "248060": {
133
+ "content": "<|fim_prefix|>",
134
+ "lstrip": false,
135
+ "normalized": false,
136
+ "rstrip": false,
137
+ "single_word": false,
138
+ "special": false
139
+ },
140
+ "248061": {
141
+ "content": "<|fim_middle|>",
142
+ "lstrip": false,
143
+ "normalized": false,
144
+ "rstrip": false,
145
+ "single_word": false,
146
+ "special": false
147
+ },
148
+ "248062": {
149
+ "content": "<|fim_suffix|>",
150
+ "lstrip": false,
151
+ "normalized": false,
152
+ "rstrip": false,
153
+ "single_word": false,
154
+ "special": false
155
+ },
156
+ "248063": {
157
+ "content": "<|fim_pad|>",
158
+ "lstrip": false,
159
+ "normalized": false,
160
+ "rstrip": false,
161
+ "single_word": false,
162
+ "special": false
163
+ },
164
+ "248064": {
165
+ "content": "<|repo_name|>",
166
+ "lstrip": false,
167
+ "normalized": false,
168
+ "rstrip": false,
169
+ "single_word": false,
170
+ "special": false
171
+ },
172
+ "248065": {
173
+ "content": "<|file_sep|>",
174
+ "lstrip": false,
175
+ "normalized": false,
176
+ "rstrip": false,
177
+ "single_word": false,
178
+ "special": false
179
+ },
180
+ "248066": {
181
+ "content": "<tool_response>",
182
+ "lstrip": false,
183
+ "normalized": false,
184
+ "rstrip": false,
185
+ "single_word": false,
186
+ "special": false
187
+ },
188
+ "248067": {
189
+ "content": "</tool_response>",
190
+ "lstrip": false,
191
+ "normalized": false,
192
+ "rstrip": false,
193
+ "single_word": false,
194
+ "special": false
195
+ },
196
+ "248068": {
197
+ "content": "<think>",
198
+ "lstrip": false,
199
+ "normalized": false,
200
+ "rstrip": false,
201
+ "single_word": false,
202
+ "special": false
203
+ },
204
+ "248069": {
205
+ "content": "</think>",
206
+ "lstrip": false,
207
+ "normalized": false,
208
+ "rstrip": false,
209
+ "single_word": false,
210
+ "special": false
211
+ },
212
+ "248070": {
213
+ "content": "<|audio_start|>",
214
+ "lstrip": false,
215
+ "normalized": false,
216
+ "rstrip": false,
217
+ "single_word": false,
218
+ "special": true
219
+ },
220
+ "248071": {
221
+ "content": "<|audio_end|>",
222
+ "lstrip": false,
223
+ "normalized": false,
224
+ "rstrip": false,
225
+ "single_word": false,
226
+ "special": true
227
+ },
228
+ "248072": {
229
+ "content": "<tts_pad>",
230
+ "lstrip": false,
231
+ "normalized": false,
232
+ "rstrip": false,
233
+ "single_word": false,
234
+ "special": true
235
+ },
236
+ "248073": {
237
+ "content": "<tts_text_bos>",
238
+ "lstrip": false,
239
+ "normalized": false,
240
+ "rstrip": false,
241
+ "single_word": false,
242
+ "special": true
243
+ },
244
+ "248074": {
245
+ "content": "<tts_text_eod>",
246
+ "lstrip": false,
247
+ "normalized": false,
248
+ "rstrip": false,
249
+ "single_word": false,
250
+ "special": true
251
+ },
252
+ "248075": {
253
+ "content": "<tts_text_bos_single>",
254
+ "lstrip": false,
255
+ "normalized": false,
256
+ "rstrip": false,
257
+ "single_word": false,
258
+ "special": true
259
+ },
260
+ "248076": {
261
+ "content": "<|audio_pad|>",
262
+ "lstrip": false,
263
+ "normalized": false,
264
+ "rstrip": false,
265
+ "single_word": false,
266
+ "special": true
267
+ }
268
+ },
269
+ "additional_special_tokens": [
270
+ "<|im_start|>",
271
+ "<|im_end|>",
272
+ "<|object_ref_start|>",
273
+ "<|object_ref_end|>",
274
+ "<|box_start|>",
275
+ "<|box_end|>",
276
+ "<|quad_start|>",
277
+ "<|quad_end|>",
278
+ "<|vision_start|>",
279
+ "<|vision_end|>",
280
+ "<|vision_pad|>",
281
+ "<|image_pad|>",
282
+ "<|video_pad|>"
283
+ ],
284
+ "bos_token": null,
285
+ "chat_template": "{%- set image_count = namespace(value=0) %}\n{%- set video_count = namespace(value=0) %}\n{%- macro render_content(content, do_vision_count, is_system_content=false) %}\n {%- if content is string %}\n {{- content }}\n {%- elif content is iterable and content is not mapping %}\n {%- for item in content %}\n {%- if 'image' in item or 'image_url' in item or item.type == 'image' %}\n {%- if is_system_content %}\n {{- raise_exception('System message cannot contain images.') }}\n {%- endif %}\n {%- if do_vision_count %}\n {%- set image_count.value = image_count.value + 1 %}\n {%- endif %}\n {%- if add_vision_id %}\n {{- 'Picture ' ~ image_count.value ~ ': ' }}\n {%- endif %}\n {{- '<|vision_start|><|image_pad|><|vision_end|>' }}\n {%- elif 'video' in item or item.type == 'video' %}\n {%- if is_system_content %}\n {{- raise_exception('System message cannot contain videos.') }}\n {%- endif %}\n {%- if do_vision_count %}\n {%- set video_count.value = video_count.value + 1 %}\n {%- endif %}\n {%- if add_vision_id %}\n {{- 'Video ' ~ video_count.value ~ ': ' }}\n {%- endif %}\n {{- '<|vision_start|><|video_pad|><|vision_end|>' }}\n {%- elif 'text' in item %}\n {{- item.text }}\n {%- else %}\n {{- raise_exception('Unexpected item type in content.') }}\n {%- endif %}\n {%- endfor %}\n {%- elif content is none or content is undefined %}\n {{- '' }}\n {%- else %}\n {{- raise_exception('Unexpected content type.') }}\n {%- endif %}\n{%- endmacro %}\n{%- if not messages %}\n {{- raise_exception('No messages provided.') }}\n{%- endif %}\n{%- if tools and tools is iterable and tools is not mapping %}\n {{- '<|im_start|>system\\n' }}\n {{- \"# Tools\\n\\nYou have access to the following functions:\\n\\n<tools>\" }}\n {%- for tool in tools %}\n {{- \"\\n\" }}\n {{- tool | tojson }}\n {%- endfor %}\n {{- \"\\n</tools>\" }}\n {{- '\\n\\nIf you choose to call a function ONLY reply in the following format with NO suffix:\\n\\n<tool_call>\\n<function=example_function_name>\\n<parameter=example_parameter_1>\\nvalue_1\\n</parameter>\\n<parameter=example_parameter_2>\\nThis is the value for the second parameter\\nthat can span\\nmultiple lines\\n</parameter>\\n</function>\\n</tool_call>\\n\\n<IMPORTANT>\\nReminder:\\n- Function calls MUST follow the specified format: an inner <function=...></function> block must be nested within <tool_call></tool_call> XML tags\\n- Required parameters MUST be specified\\n- You may provide optional reasoning for your function call in natural language BEFORE the function call, but NOT after\\n- If there is no function call available, answer the question like normal with your current knowledge and do not tell the user about function calls\\n</IMPORTANT>' }}\n {%- if messages[0].role == 'system' %}\n {%- set content = render_content(messages[0].content, false, true)|trim %}\n {%- if content %}\n {{- '\\n\\n' + content }}\n {%- endif %}\n {%- endif %}\n {{- '<|im_end|>\\n' }}\n{%- else %}\n {%- if messages[0].role == 'system' %}\n {%- set content = render_content(messages[0].content, false, true)|trim %}\n {{- '<|im_start|>system\\n' + content + '<|im_end|>\\n' }}\n {%- endif %}\n{%- endif %}\n{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}\n{%- for message in messages[::-1] %}\n {%- set index = (messages|length - 1) - loop.index0 %}\n {%- if ns.multi_step_tool and message.role == \"user\" %}\n {%- set content = render_content(message.content, false)|trim %}\n {%- if not(content.startswith('<tool_response>') and content.endswith('</tool_response>')) %}\n {%- set ns.multi_step_tool = false %}\n {%- set ns.last_query_index = index %}\n {%- endif %}\n {%- endif %}\n{%- endfor %}\n{%- if ns.multi_step_tool %}\n {{- raise_exception('No user query found in messages.') }}\n{%- endif %}\n{%- for message in messages %}\n {%- set content = render_content(message.content, true)|trim %}\n {%- if message.role == \"system\" %}\n {%- if not loop.first %}\n {{- raise_exception('System message must be at the beginning.') }}\n {%- endif %}\n {%- elif message.role == \"user\" %}\n {{- '<|im_start|>' + message.role + '\\n' + content + '<|im_end|>' + '\\n' }}\n {%- elif message.role == \"assistant\" %}\n {%- set reasoning_content = '' %}\n {%- if message.reasoning_content is string %}\n {%- set reasoning_content = message.reasoning_content %}\n {%- else %}\n {%- if '</think>' in content %}\n {%- set reasoning_content = content.split('</think>')[0].rstrip('\\n').split('<think>')[-1].lstrip('\\n') %}\n {%- set content = content.split('</think>')[-1].lstrip('\\n') %}\n {%- endif %}\n {%- endif %}\n {%- set reasoning_content = reasoning_content|trim %}\n {%- if (preserve_thinking is defined and preserve_thinking is true) or (loop.index0 > ns.last_query_index) %}\n {{- '<|im_start|>' + message.role + '\\n<think>\\n' + reasoning_content + '\\n</think>\\n\\n' + content }}\n {%- else %}\n {{- '<|im_start|>' + message.role + '\\n' + content }}\n {%- endif %}\n {%- if message.tool_calls and message.tool_calls is iterable and message.tool_calls is not mapping %}\n {%- for tool_call in message.tool_calls %}\n {%- if tool_call.function is defined %}\n {%- set tool_call = tool_call.function %}\n {%- endif %}\n {%- if loop.first %}\n {%- if content|trim %}\n {{- '\\n\\n<tool_call>\\n<function=' + tool_call.name + '>\\n' }}\n {%- else %}\n {{- '<tool_call>\\n<function=' + tool_call.name + '>\\n' }}\n {%- endif %}\n {%- else %}\n {{- '\\n<tool_call>\\n<function=' + tool_call.name + '>\\n' }}\n {%- endif %}\n {%- if tool_call.arguments is defined %}\n {%- for args_name, args_value in tool_call.arguments|items %}\n {{- '<parameter=' + args_name + '>\\n' }}\n {%- set args_value = args_value | string if args_value is string else args_value | tojson | safe %}\n {{- args_value }}\n {{- '\\n</parameter>\\n' }}\n {%- endfor %}\n {%- endif %}\n {{- '</function>\\n</tool_call>' }}\n {%- endfor %}\n {%- endif %}\n {{- '<|im_end|>\\n' }}\n {%- elif message.role == \"tool\" %}\n {%- if loop.previtem and loop.previtem.role != \"tool\" %}\n {{- '<|im_start|>user' }}\n {%- endif %}\n {{- '\\n<tool_response>\\n' }}\n {{- content }}\n {{- '\\n</tool_response>' }}\n {%- if not loop.last and loop.nextitem.role != \"tool\" %}\n {{- '<|im_end|>\\n' }}\n {%- elif loop.last %}\n {{- '<|im_end|>\\n' }}\n {%- endif %}\n {%- else %}\n {{- raise_exception('Unexpected message role.') }}\n {%- endif %}\n{%- endfor %}\n{%- if add_generation_prompt %}\n {{- '<|im_start|>assistant\\n' }}\n {%- if enable_thinking is defined and enable_thinking is false %}\n {{- '<think>\\n\\n</think>\\n\\n' }}\n {%- else %}\n {{- '<think>\\n' }}\n {%- endif %}\n{%- endif %}",
286
+ "clean_up_tokenization_spaces": false,
287
+ "eos_token": "<|im_end|>",
288
+ "errors": "replace",
289
+ "model_max_length": 262144,
290
+ "pad_token": "<|endoftext|>",
291
+ "split_special_tokens": false,
292
+ "tokenizer_class": "Qwen2Tokenizer",
293
+ "unk_token": null,
294
+ "add_bos_token": false,
295
+ "pretokenize_regex": "(?i:'s|'t|'re|'ve|'m|'ll|'d)|[^\\r\\n\\p{L}\\p{N}]?[\\p{L}\\p{M}]+|\\p{N}| ?[^\\s\\p{L}\\p{M}\\p{N}]+[\\r\\n]*|\\s*[\\r\\n]+|\\s+(?!\\S)|\\s+",
296
+ "extra_special_tokens": {
297
+ "audio_bos_token": "<|audio_start|>",
298
+ "audio_eos_token": "<|audio_end|>",
299
+ "audio_token": "<|audio_pad|>",
300
+ "image_token": "<|image_pad|>",
301
+ "video_token": "<|video_pad|>",
302
+ "vision_bos_token": "<|vision_start|>",
303
+ "vision_eos_token": "<|vision_end|>"
304
+ }
305
+ }