Hugging Face's logo

Urabewe
/
Ace-Step-Captioner-4bit

Safetensors
qwen2_5_omni
4-bit precision
bitsandbytes
Model card Files
xet
Community
Urabewe commited on
Commit
f6f3f55
·
verified ·
1 Parent(s): 9ab57a6

Upload 7 files

Browse files
Files changed (8) hide show
  1. .gitattributes +1 -0
  2. chat_template.jinja +7 -0
  3. config.json +529 -0
  4. generation_config.json +8 -0
  5. model.safetensors +3 -0
  6. processor_config.json +116 -0
  7. tokenizer.json +3 -0
  8. tokenizer_config.json +31 -0
.gitattributes CHANGED
@@ -33,3 +33,4 @@ 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
+ tokenizer.json filter=lfs diff=lfs merge=lfs -text
chat_template.jinja ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {% set audio_count = namespace(value=0) %}{% set image_count = namespace(value=0) %}{% set video_count = namespace(value=0) %}{% for message in messages %}{% if loop.first and message['role'] != 'system' %}<|im_start|>system
2
+ You are a helpful assistant.<|im_end|>
3
+ {% endif %}<|im_start|>{{ message['role'] }}
4
+ {% if message['content'] is string %}{{ message['content'] }}<|im_end|>
5
+ {% else %}{% for content in message['content'] %}{% if content['type'] == 'image' or 'image' in content or 'image_url' in content %}{% set image_count.value = image_count.value + 1 %}{% if add_vision_id %}Picture {{ image_count.value }}: {% endif %}<|vision_bos|><|IMAGE|><|vision_eos|>{% elif content['type'] == 'audio' or 'audio' in content or 'audio_url' in content %}{% set audio_count.value = audio_count.value + 1 %}{% if add_audio_id %}Audio {{ audio_count.value }}: {% endif %}<|audio_bos|><|AUDIO|><|audio_eos|>{% elif content['type'] == 'video' or 'video' in content %}{% set video_count.value = video_count.value + 1 %}{% if add_vision_id %}Video {{ video_count.value }}: {% endif %}<|vision_bos|><|VIDEO|><|vision_eos|>{% elif 'text' in content %}{{ content['text'] }}{% endif %}{% endfor %}<|im_end|>
6
+ {% endif %}{% endfor %}{% if add_generation_prompt %}<|im_start|>assistant
7
+ {% endif %}
config.json ADDED
@@ -0,0 +1,529 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "Qwen2_5OmniForConditionalGeneration"
4
+ ],
5
+ "dtype": "bfloat16",
6
+ "enable_audio_output": true,
7
+ "enable_talker": true,
8
+ "hidden_size": 3584,
9
+ "keys_to_ignore_at_inference": [
10
+ "past_key_values",
11
+ "hidden_states",
12
+ "attention_mask",
13
+ "hidden_states",
14
+ "attention_mask",
15
+ "hidden_states",
16
+ "attention_mask"
17
+ ],
18
+ "model_type": "qwen2_5_omni",
19
+ "pad_token_id": 151643,
20
+ "quantization_config": {
21
+ "_load_in_4bit": true,
22
+ "_load_in_8bit": false,
23
+ "bnb_4bit_compute_dtype": "bfloat16",
24
+ "bnb_4bit_quant_storage": "uint8",
25
+ "bnb_4bit_quant_type": "nf4",
26
+ "bnb_4bit_use_double_quant": true,
27
+ "llm_int8_enable_fp32_cpu_offload": false,
28
+ "llm_int8_has_fp16_weight": false,
29
+ "llm_int8_skip_modules": null,
30
+ "llm_int8_threshold": 6.0,
31
+ "load_in_4bit": true,
32
+ "load_in_8bit": false,
33
+ "quant_method": "bitsandbytes"
34
+ },
35
+ "talker_config": {
36
+ "_name_or_path": "Qwen2.5-Omni-7B/talker",
37
+ "architectures": [
38
+ "Qwen2OmniTalkerForConditionalGeneration"
39
+ ],
40
+ "attention_dropout": 0.0,
41
+ "audio_end_token_id": 151648,
42
+ "audio_start_token_id": 151647,
43
+ "audio_token_index": 151646,
44
+ "dtype": "bfloat16",
45
+ "embedding_size": 3584,
46
+ "head_dim": 128,
47
+ "hidden_act": "silu",
48
+ "hidden_size": 896,
49
+ "image_token_index": 151655,
50
+ "init_std": 0.02,
51
+ "initializer_range": 0.02,
52
+ "intermediate_size": 18944,
53
+ "layer_types": [
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
+ "full_attention",
75
+ "full_attention",
76
+ "full_attention",
77
+ "full_attention"
78
+ ],
79
+ "max_position_embeddings": 32768,
80
+ "max_window_layers": 28,
81
+ "model_type": "qwen2_5_omni_talker",
82
+ "num_attention_heads": 12,
83
+ "num_hidden_layers": 24,
84
+ "num_key_value_heads": 4,
85
+ "pad_token_id": null,
86
+ "position_id_per_seconds": 25,
87
+ "rms_norm_eps": 1e-06,
88
+ "rope_parameters": {
89
+ "mrope_section": [
90
+ 16,
91
+ 24,
92
+ 24
93
+ ],
94
+ "rope_theta": 1000000.0,
95
+ "rope_type": "default",
96
+ "type": "default"
97
+ },
98
+ "seconds_per_chunk": 2,
99
+ "sliding_window": null,
100
+ "spatial_merge_size": 2,
101
+ "tie_word_embeddings": false,
102
+ "tts_codec_end_token_id": 8294,
103
+ "tts_codec_mask_token_id": 8296,
104
+ "tts_codec_pad_token_id": 8292,
105
+ "tts_codec_start_token_id": 8293,
106
+ "tts_text_end_token_id": 151861,
107
+ "tts_text_pad_token_id": 151859,
108
+ "tts_text_start_token_id": 151860,
109
+ "use_cache": false,
110
+ "use_sliding_window": false,
111
+ "video_token_index": 151656,
112
+ "vision_end_token_id": 151653,
113
+ "vision_start_token_id": 151652,
114
+ "vocab_size": 8448
115
+ },
116
+ "thinker_config": {
117
+ "_name_or_path": "Qwen2.5-Omni-7B/thinker",
118
+ "architectures": [
119
+ "Qwen2OmniNaViTThinkerForConditionalGeneration"
120
+ ],
121
+ "audio_config": {
122
+ "_name_or_path": "",
123
+ "activation_dropout": 0.0,
124
+ "activation_function": "gelu",
125
+ "add_cross_attention": false,
126
+ "architectures": null,
127
+ "attention_dropout": 0.0,
128
+ "bos_token_id": null,
129
+ "chunk_size_feed_forward": 0,
130
+ "cross_attention_hidden_size": null,
131
+ "d_model": 1280,
132
+ "decoder_start_token_id": null,
133
+ "dropout": 0.0,
134
+ "dtype": null,
135
+ "encoder_attention_heads": 20,
136
+ "encoder_ffn_dim": 5120,
137
+ "encoder_layerdrop": 0.0,
138
+ "encoder_layers": 32,
139
+ "eos_token_id": null,
140
+ "finetuning_task": null,
141
+ "id2label": {
142
+ "0": "LABEL_0",
143
+ "1": "LABEL_1"
144
+ },
145
+ "init_std": 0.02,
146
+ "initializer_range": 0.02,
147
+ "is_decoder": false,
148
+ "is_encoder_decoder": false,
149
+ "label2id": {
150
+ "LABEL_0": 0,
151
+ "LABEL_1": 1
152
+ },
153
+ "max_source_positions": 1500,
154
+ "model_type": "qwen2_5_omni_audio_encoder",
155
+ "n_window": 100,
156
+ "num_hidden_layers": 32,
157
+ "num_mel_bins": 128,
158
+ "output_attentions": false,
159
+ "output_dim": 3584,
160
+ "output_hidden_states": false,
161
+ "pad_token_id": null,
162
+ "prefix": null,
163
+ "problem_type": null,
164
+ "pruned_heads": {},
165
+ "return_dict": true,
166
+ "scale_embedding": false,
167
+ "sep_token_id": null,
168
+ "task_specific_params": null,
169
+ "tf_legacy_loss": false,
170
+ "tie_encoder_decoder": false,
171
+ "tie_word_embeddings": true,
172
+ "tokenizer_class": null,
173
+ "torchscript": false,
174
+ "use_bfloat16": false
175
+ },
176
+ "audio_end_token_id": 151648,
177
+ "audio_start_token_id": 151647,
178
+ "audio_token_index": 151646,
179
+ "bos_token_id": 151644,
180
+ "dtype": "bfloat16",
181
+ "eos_token_id": 151645,
182
+ "ignore_index": -100,
183
+ "image_token_index": 151655,
184
+ "init_std": 0.02,
185
+ "initializer_range": 0.02,
186
+ "model_type": "qwen2_5_omni_thinker",
187
+ "pad_token_id": 151643,
188
+ "position_id_per_seconds": 25,
189
+ "seconds_per_chunk": 2,
190
+ "text_config": {
191
+ "_name_or_path": "",
192
+ "add_cross_attention": false,
193
+ "architectures": null,
194
+ "attention_dropout": 0.0,
195
+ "bos_token_id": null,
196
+ "chunk_size_feed_forward": 0,
197
+ "cross_attention_hidden_size": null,
198
+ "decoder_start_token_id": null,
199
+ "dtype": null,
200
+ "eos_token_id": null,
201
+ "finetuning_task": null,
202
+ "hidden_act": "silu",
203
+ "hidden_size": 3584,
204
+ "id2label": {
205
+ "0": "LABEL_0",
206
+ "1": "LABEL_1"
207
+ },
208
+ "init_std": 0.02,
209
+ "initializer_range": 0.02,
210
+ "intermediate_size": 18944,
211
+ "is_decoder": false,
212
+ "is_encoder_decoder": false,
213
+ "label2id": {
214
+ "LABEL_0": 0,
215
+ "LABEL_1": 1
216
+ },
217
+ "layer_types": [
218
+ "full_attention",
219
+ "full_attention",
220
+ "full_attention",
221
+ "full_attention",
222
+ "full_attention",
223
+ "full_attention",
224
+ "full_attention",
225
+ "full_attention",
226
+ "full_attention",
227
+ "full_attention",
228
+ "full_attention",
229
+ "full_attention",
230
+ "full_attention",
231
+ "full_attention",
232
+ "full_attention",
233
+ "full_attention",
234
+ "full_attention",
235
+ "full_attention",
236
+ "full_attention",
237
+ "full_attention",
238
+ "full_attention",
239
+ "full_attention",
240
+ "full_attention",
241
+ "full_attention",
242
+ "full_attention",
243
+ "full_attention",
244
+ "full_attention",
245
+ "full_attention"
246
+ ],
247
+ "max_position_embeddings": 32768,
248
+ "max_window_layers": 28,
249
+ "model_type": "qwen2_5_omni_text",
250
+ "num_attention_heads": 28,
251
+ "num_hidden_layers": 28,
252
+ "num_key_value_heads": 4,
253
+ "output_attentions": false,
254
+ "output_hidden_states": false,
255
+ "pad_token_id": null,
256
+ "prefix": null,
257
+ "problem_type": null,
258
+ "pruned_heads": {},
259
+ "return_dict": true,
260
+ "rms_norm_eps": 1e-06,
261
+ "rope_parameters": {
262
+ "mrope_section": [
263
+ 16,
264
+ 24,
265
+ 24
266
+ ],
267
+ "rope_theta": 1000000.0,
268
+ "rope_type": "default",
269
+ "type": "default"
270
+ },
271
+ "sep_token_id": null,
272
+ "sliding_window": null,
273
+ "task_specific_params": null,
274
+ "tf_legacy_loss": false,
275
+ "tie_encoder_decoder": false,
276
+ "tie_word_embeddings": false,
277
+ "tokenizer_class": null,
278
+ "torchscript": false,
279
+ "use_bfloat16": false,
280
+ "use_cache": true,
281
+ "use_sliding_window": false,
282
+ "vocab_size": 152064
283
+ },
284
+ "tie_word_embeddings": false,
285
+ "user_token_id": 872,
286
+ "video_token_index": 151656,
287
+ "vision_config": {
288
+ "_name_or_path": "",
289
+ "add_cross_attention": false,
290
+ "architectures": null,
291
+ "bos_token_id": null,
292
+ "chunk_size_feed_forward": 0,
293
+ "cross_attention_hidden_size": null,
294
+ "decoder_start_token_id": null,
295
+ "depth": 32,
296
+ "dtype": null,
297
+ "embed_dim": 1280,
298
+ "eos_token_id": null,
299
+ "finetuning_task": null,
300
+ "fullatt_block_indexes": [
301
+ 7,
302
+ 15,
303
+ 23,
304
+ 31
305
+ ],
306
+ "hidden_act": "silu",
307
+ "hidden_size": 1280,
308
+ "id2label": {
309
+ "0": "LABEL_0",
310
+ "1": "LABEL_1"
311
+ },
312
+ "in_channels": 3,
313
+ "in_chans": 3,
314
+ "init_std": 0.02,
315
+ "initializer_range": 0.02,
316
+ "intermediate_size": 3420,
317
+ "is_decoder": false,
318
+ "is_encoder_decoder": false,
319
+ "label2id": {
320
+ "LABEL_0": 0,
321
+ "LABEL_1": 1
322
+ },
323
+ "model_type": "qwen2_5_omni_vision_encoder",
324
+ "num_heads": 16,
325
+ "out_hidden_size": 3584,
326
+ "output_attentions": false,
327
+ "output_hidden_states": false,
328
+ "pad_token_id": null,
329
+ "patch_size": 14,
330
+ "prefix": null,
331
+ "problem_type": null,
332
+ "pruned_heads": {},
333
+ "return_dict": true,
334
+ "sep_token_id": null,
335
+ "spatial_merge_size": 2,
336
+ "spatial_patch_size": 14,
337
+ "task_specific_params": null,
338
+ "temporal_patch_size": 2,
339
+ "tf_legacy_loss": false,
340
+ "tie_encoder_decoder": false,
341
+ "tie_word_embeddings": true,
342
+ "tokenizer_class": null,
343
+ "tokens_per_second": 25,
344
+ "torchscript": false,
345
+ "use_bfloat16": false,
346
+ "window_size": 112
347
+ },
348
+ "vision_end_token_id": 151653,
349
+ "vision_start_token_id": 151652,
350
+ "vision_token_id": 151654
351
+ },
352
+ "token2wav_config": {
353
+ "bigvgan_config": {
354
+ "_name_or_path": "",
355
+ "add_cross_attention": false,
356
+ "architectures": null,
357
+ "bos_token_id": null,
358
+ "chunk_size_feed_forward": 0,
359
+ "cross_attention_hidden_size": null,
360
+ "decoder_start_token_id": null,
361
+ "dtype": null,
362
+ "eos_token_id": null,
363
+ "finetuning_task": null,
364
+ "id2label": {
365
+ "0": "LABEL_0",
366
+ "1": "LABEL_1"
367
+ },
368
+ "is_decoder": false,
369
+ "is_encoder_decoder": false,
370
+ "label2id": {
371
+ "LABEL_0": 0,
372
+ "LABEL_1": 1
373
+ },
374
+ "mel_dim": 80,
375
+ "model_type": "qwen2_5_omni_bigvgan",
376
+ "output_attentions": false,
377
+ "output_hidden_states": false,
378
+ "pad_token_id": null,
379
+ "prefix": null,
380
+ "problem_type": null,
381
+ "pruned_heads": {},
382
+ "resblock_dilation_sizes": [
383
+ [
384
+ 1,
385
+ 3,
386
+ 5
387
+ ],
388
+ [
389
+ 1,
390
+ 3,
391
+ 5
392
+ ],
393
+ [
394
+ 1,
395
+ 3,
396
+ 5
397
+ ]
398
+ ],
399
+ "resblock_kernel_sizes": [
400
+ 3,
401
+ 7,
402
+ 11
403
+ ],
404
+ "return_dict": true,
405
+ "sep_token_id": null,
406
+ "task_specific_params": null,
407
+ "tf_legacy_loss": false,
408
+ "tie_encoder_decoder": false,
409
+ "tie_word_embeddings": true,
410
+ "tokenizer_class": null,
411
+ "torchscript": false,
412
+ "upsample_initial_channel": 1536,
413
+ "upsample_kernel_sizes": [
414
+ 11,
415
+ 7,
416
+ 4,
417
+ 4,
418
+ 4,
419
+ 4
420
+ ],
421
+ "upsample_rates": [
422
+ 5,
423
+ 3,
424
+ 2,
425
+ 2,
426
+ 2,
427
+ 2
428
+ ],
429
+ "use_bfloat16": false,
430
+ "use_bias_at_final": false
431
+ },
432
+ "dit_config": {
433
+ "_name_or_path": "",
434
+ "add_cross_attention": false,
435
+ "architectures": null,
436
+ "block_size": 24,
437
+ "bos_token_id": null,
438
+ "chunk_size_feed_forward": 0,
439
+ "cross_attention_hidden_size": null,
440
+ "decoder_start_token_id": null,
441
+ "depth": 22,
442
+ "dim": 1024,
443
+ "dropout": 0.1,
444
+ "dtype": "float32",
445
+ "emb_dim": 512,
446
+ "enc_attention_channels": 64,
447
+ "enc_channels": [
448
+ 256,
449
+ 256,
450
+ 256,
451
+ 256,
452
+ 768
453
+ ],
454
+ "enc_dilations": [
455
+ 1,
456
+ 2,
457
+ 3,
458
+ 4,
459
+ 1
460
+ ],
461
+ "enc_dim": 128,
462
+ "enc_emb_dim": 192,
463
+ "enc_global_context": true,
464
+ "enc_kernel_sizes": [
465
+ 5,
466
+ 3,
467
+ 3,
468
+ 3,
469
+ 1
470
+ ],
471
+ "enc_lin_neurons": 192,
472
+ "enc_res2net_scale": 2,
473
+ "enc_se_channels": 64,
474
+ "eos_token_id": null,
475
+ "ff_mult": 2,
476
+ "finetuning_task": null,
477
+ "head_dim": 64,
478
+ "heads": 16,
479
+ "hidden_size": 1024,
480
+ "id2label": {
481
+ "0": "LABEL_0",
482
+ "1": "LABEL_1"
483
+ },
484
+ "is_decoder": false,
485
+ "is_encoder_decoder": false,
486
+ "label2id": {
487
+ "LABEL_0": 0,
488
+ "LABEL_1": 1
489
+ },
490
+ "look_ahead_layers": [
491
+ 10
492
+ ],
493
+ "look_backward_layers": [
494
+ 0,
495
+ 20
496
+ ],
497
+ "max_position_embeddings": 32768,
498
+ "mel_dim": 80,
499
+ "model_type": "qwen2_5_omni_dit",
500
+ "num_attention_heads": 16,
501
+ "num_embeds": 8193,
502
+ "num_hidden_layers": 22,
503
+ "output_attentions": false,
504
+ "output_hidden_states": false,
505
+ "pad_token_id": null,
506
+ "prefix": null,
507
+ "problem_type": null,
508
+ "pruned_heads": {},
509
+ "repeats": 2,
510
+ "return_dict": true,
511
+ "rope_parameters": {
512
+ "rope_theta": 10000.0,
513
+ "rope_type": "default"
514
+ },
515
+ "sep_token_id": null,
516
+ "task_specific_params": null,
517
+ "tf_legacy_loss": false,
518
+ "tie_encoder_decoder": false,
519
+ "tie_word_embeddings": true,
520
+ "tokenizer_class": null,
521
+ "torchscript": false,
522
+ "use_bfloat16": false
523
+ },
524
+ "dtype": "bfloat16",
525
+ "model_type": "qwen2_5_omni_token2wav",
526
+ "pad_token_id": 151643
527
+ },
528
+ "transformers_version": "5.1.0"
529
+ }
generation_config.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "eos_token_id": [
4
+ 151645,
5
+ 151643
6
+ ],
7
+ "transformers_version": "5.1.0"
8
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:75e3bf91ddbc2e8ab66a5f84300102fda1b5e281ec7fd5f16b89e25d5fc6940e
3
+ size 6828061916
processor_config.json ADDED
@@ -0,0 +1,116 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "feature_extractor": {
3
+ "chunk_length": 300,
4
+ "dither": 0.0,
5
+ "feature_extractor_type": "WhisperFeatureExtractor",
6
+ "feature_size": 128,
7
+ "hop_length": 160,
8
+ "image_mean": [
9
+ 0.48145466,
10
+ 0.4578275,
11
+ 0.40821073
12
+ ],
13
+ "image_processor_type": "Qwen2VLImageProcessor",
14
+ "image_std": [
15
+ 0.26862954,
16
+ 0.26130258,
17
+ 0.27577711
18
+ ],
19
+ "max_pixels": 12845056,
20
+ "merge_size": 2,
21
+ "min_pixels": 3136,
22
+ "n_fft": 400,
23
+ "n_samples": 4800000,
24
+ "nb_max_frames": 30000,
25
+ "padding_side": "right",
26
+ "padding_value": 0.0,
27
+ "patch_size": 14,
28
+ "return_attention_mask": true,
29
+ "sampling_rate": 16000,
30
+ "temporal_patch_size": 2
31
+ },
32
+ "image_processor": {
33
+ "chunk_length": 300,
34
+ "data_format": "channels_first",
35
+ "dither": 0.0,
36
+ "do_convert_rgb": true,
37
+ "do_normalize": true,
38
+ "do_rescale": true,
39
+ "do_resize": true,
40
+ "feature_size": 128,
41
+ "hop_length": 160,
42
+ "image_mean": [
43
+ 0.48145466,
44
+ 0.4578275,
45
+ 0.40821073
46
+ ],
47
+ "image_processor_type": "Qwen2VLImageProcessorFast",
48
+ "image_std": [
49
+ 0.26862954,
50
+ 0.26130258,
51
+ 0.27577711
52
+ ],
53
+ "merge_size": 2,
54
+ "n_fft": 400,
55
+ "n_samples": 4800000,
56
+ "nb_max_frames": 30000,
57
+ "padding_side": "right",
58
+ "padding_value": 0.0,
59
+ "patch_size": 14,
60
+ "resample": 3,
61
+ "rescale_factor": 0.00392156862745098,
62
+ "return_attention_mask": true,
63
+ "sampling_rate": 16000,
64
+ "size": {
65
+ "longest_edge": 12845056,
66
+ "shortest_edge": 3136
67
+ },
68
+ "temporal_patch_size": 2
69
+ },
70
+ "processor_class": "Qwen2_5OmniProcessor",
71
+ "video_processor": {
72
+ "chunk_length": 300,
73
+ "data_format": "channels_first",
74
+ "default_to_square": true,
75
+ "dither": 0.0,
76
+ "do_convert_rgb": true,
77
+ "do_normalize": true,
78
+ "do_rescale": true,
79
+ "do_resize": true,
80
+ "do_sample_frames": false,
81
+ "feature_extractor_type": "WhisperFeatureExtractor",
82
+ "feature_size": 128,
83
+ "hop_length": 160,
84
+ "image_mean": [
85
+ 0.48145466,
86
+ 0.4578275,
87
+ 0.40821073
88
+ ],
89
+ "image_processor_type": "Qwen2VLImageProcessor",
90
+ "image_std": [
91
+ 0.26862954,
92
+ 0.26130258,
93
+ 0.27577711
94
+ ],
95
+ "max_frames": 768,
96
+ "merge_size": 2,
97
+ "min_frames": 4,
98
+ "n_fft": 400,
99
+ "n_samples": 4800000,
100
+ "nb_max_frames": 30000,
101
+ "padding_side": "right",
102
+ "padding_value": 0.0,
103
+ "patch_size": 14,
104
+ "resample": 3,
105
+ "rescale_factor": 0.00392156862745098,
106
+ "return_attention_mask": true,
107
+ "return_metadata": false,
108
+ "sampling_rate": 16000,
109
+ "size": {
110
+ "longest_edge": 12845056,
111
+ "shortest_edge": 3136
112
+ },
113
+ "temporal_patch_size": 2,
114
+ "video_processor_type": "Qwen2VLVideoProcessor"
115
+ }
116
+ }
tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1ab7a851e5c63d5fafbfdac72e3b4a8d08613f6bbb06ee39036fdf870ef59c93
3
+ size 11421866
tokenizer_config.json ADDED
@@ -0,0 +1,31 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "audio_bos_token": "<|audio_bos|>",
4
+ "audio_eos_token": "<|audio_eos|>",
5
+ "audio_token": "<|AUDIO|>",
6
+ "backend": "tokenizers",
7
+ "bos_token": null,
8
+ "clean_up_tokenization_spaces": false,
9
+ "eos_token": "<|im_end|>",
10
+ "errors": "replace",
11
+ "image_token": "<|IMAGE|>",
12
+ "is_local": true,
13
+ "model_max_length": 32768,
14
+ "model_specific_special_tokens": {
15
+ "audio_bos_token": "<|audio_bos|>",
16
+ "audio_eos_token": "<|audio_eos|>",
17
+ "audio_token": "<|AUDIO|>",
18
+ "image_token": "<|IMAGE|>",
19
+ "video_token": "<|VIDEO|>",
20
+ "vision_bos_token": "<|vision_bos|>",
21
+ "vision_eos_token": "<|vision_eos|>"
22
+ },
23
+ "pad_token": "<|endoftext|>",
24
+ "processor_class": "Qwen2_5OmniProcessor",
25
+ "split_special_tokens": false,
26
+ "tokenizer_class": "Qwen2Tokenizer",
27
+ "unk_token": null,
28
+ "video_token": "<|VIDEO|>",
29
+ "vision_bos_token": "<|vision_bos|>",
30
+ "vision_eos_token": "<|vision_eos|>"
31
+ }