Hugging Face's logo

Parabridge
/
MainCheckpoint

Safetensors
qwen3_omni_moe
Model card Files
xet
 Community
Parabridge commited on
Commit
6f5bbda
·
verified ·
1 Parent(s): 29ba6cf

Add files using upload-large-folder tool

Browse files
Files changed (26) hide show
  1. .gitattributes +1 -0
  2. added_tokens.json +35 -0
  3. chat_template.jinja +105 -0
  4. config.json +524 -0
  5. generation_config.json +8 -0
  6. merges.txt +0 -0
  7. model-00001-of-00013.safetensors +3 -0
  8. model-00002-of-00013.safetensors +3 -0
  9. model-00003-of-00013.safetensors +3 -0
  10. model-00004-of-00013.safetensors +3 -0
  11. model-00005-of-00013.safetensors +3 -0
  12. model-00006-of-00013.safetensors +3 -0
  13. model-00007-of-00013.safetensors +3 -0
  14. model-00008-of-00013.safetensors +3 -0
  15. model-00009-of-00013.safetensors +3 -0
  16. model-00010-of-00013.safetensors +3 -0
  17. model-00011-of-00013.safetensors +3 -0
  18. model-00012-of-00013.safetensors +3 -0
  19. model-00013-of-00013.safetensors +3 -0
  20. model.safetensors.index.json +0 -0
  21. preprocessor_config.json +31 -0
  22. special_tokens_map.json +44 -0
  23. tokenizer.json +3 -0
  24. tokenizer_config.json +317 -0
  25. video_preprocessor_config.json +54 -0
  26. vocab.json +0 -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
added_tokens.json ADDED
@@ -0,0 +1,35 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "</think>": 151668,
3
+ "</tool_call>": 151658,
4
+ "</tool_response>": 151666,
5
+ "<think>": 151667,
6
+ "<tool_call>": 151657,
7
+ "<tool_response>": 151665,
8
+ "<tts_pad>": 151671,
9
+ "<tts_text_bos>": 151672,
10
+ "<tts_text_bos_single>": 151674,
11
+ "<tts_text_eod>": 151673,
12
+ "<|audio_end|>": 151670,
13
+ "<|audio_pad|>": 151675,
14
+ "<|audio_start|>": 151669,
15
+ "<|box_end|>": 151649,
16
+ "<|box_start|>": 151648,
17
+ "<|endoftext|>": 151643,
18
+ "<|file_sep|>": 151664,
19
+ "<|fim_middle|>": 151660,
20
+ "<|fim_pad|>": 151662,
21
+ "<|fim_prefix|>": 151659,
22
+ "<|fim_suffix|>": 151661,
23
+ "<|im_end|>": 151645,
24
+ "<|im_start|>": 151644,
25
+ "<|image_pad|>": 151655,
26
+ "<|object_ref_end|>": 151647,
27
+ "<|object_ref_start|>": 151646,
28
+ "<|quad_end|>": 151651,
29
+ "<|quad_start|>": 151650,
30
+ "<|repo_name|>": 151663,
31
+ "<|video_pad|>": 151656,
32
+ "<|vision_end|>": 151653,
33
+ "<|vision_pad|>": 151654,
34
+ "<|vision_start|>": 151652
35
+ }
chat_template.jinja ADDED
@@ -0,0 +1,105 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {%- if tools %}
2
+ {{- '<|im_start|>system\n' }}
3
+ {%- if messages[0].role == 'system' %}{{- messages[0].content + '\n\n' }}{%- endif %}
4
+ {{- "# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
5
+ {%- for tool in tools %}
6
+ {{- "\n" }}
7
+ {{- tool | tojson }}
8
+ {%- endfor %}
9
+ {{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
10
+ {%- else %}
11
+ {%- if messages[0].role == 'system' %}
12
+ {%- if messages[0].content is string %}
13
+ {{- '<|im_start|>system\n' + messages[0].content + '<|im_end|>\n' }}
14
+ {%- else %}
15
+ {%- for content in messages[0].content %}
16
+ {%- if content.type == 'image' or 'image' in content or 'image_url' in content %}
17
+ {{- '<|im_start|>system\n' +"<|vision_start|><|image_pad|><|vision_end|>"+ '<|im_end|>\n' }}
18
+ {%- elif content.type == 'audio' or 'audio' in content or 'audio_url' in content %}
19
+ {{- '<|im_start|>system\n' +"<|audio_start|><|audio_pad|><|audio_end|>"+ '<|im_end|>\n' }}
20
+ {%- elif content.type == 'video' or 'video' in content %}
21
+ {{- '<|im_start|>system\n' +"<|vision_start|><|video_pad|><|vision_end|>"+ '<|im_end|>\n' }}
22
+ {%- elif content.type == 'text' %}
23
+ {{- '<|im_start|>system\n' +content.text+ '<|im_end|>\n' }}
24
+ {%- endif %}
25
+ {%- endfor %}
26
+ {%- endif %}
27
+ {%- endif %}
28
+ {%- endif %}
29
+ {%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
30
+ {%- for message in messages[::-1] %}
31
+ {%- set index = (messages|length - 1) - loop.index0 %}
32
+ {%- if ns.multi_step_tool and message.role == "user" and message.content is string and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}
33
+ {%- set ns.multi_step_tool = false %}
34
+ {%- set ns.last_query_index = index %}
35
+ {%- endif %}
36
+ {%- endfor %}
37
+ {%- for message in messages %}
38
+ {%- if message.content is string %}
39
+ {%- set content = message.content %}
40
+ {%- else %}
41
+ {%- set content = namespace(text="") %}
42
+ {%- for mcontent in message.content %}
43
+ {%- if mcontent.type == 'image' or 'image' in mcontent or 'image_url' in mcontent %}
44
+ {%- set content.text = content.text~"<|vision_start|><|image_pad|><|vision_end|>" %}
45
+ {%- elif mcontent.type == 'audio' or 'audio' in mcontent or 'audio_url' in mcontent %}
46
+ {%- set content.text = content.text~"<|audio_start|><|audio_pad|><|audio_end|>" %}
47
+ {%- elif mcontent.type == 'video' or 'video' in mcontent %}
48
+ {%- set content.text = content.text~"<|vision_start|><|video_pad|><|vision_end|>" %}
49
+ {%- elif mcontent.type == 'text' %}
50
+ {%- set content.text = content.text~mcontent.text %}
51
+ {%- endif %}
52
+ {%- endfor %}
53
+ {%- set content = content.text %}
54
+ {%- endif %}
55
+ {%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
56
+ {{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
57
+ {%- elif message.role == "assistant" %}
58
+ {%- set reasoning_content = "" %}
59
+ {%- if message.reasoning_content is string %}
60
+ {%- set reasoning_content = message.reasoning_content %}
61
+ {%- else %}
62
+ {%- if '</think>' in content %}
63
+ {%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
64
+ {%- set content = content.split('</think>')[-1].lstrip('\n') %}
65
+ {%- endif %}
66
+ {%- endif %}
67
+ {%- if loop.index0 > ns.last_query_index %}
68
+ {%- if loop.last or (not loop.last and reasoning_content) %}
69
+ {{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content.strip("\n") + '\n</think>\n\n' + content.lstrip('\n') }}
70
+ {%- else %}
71
+ {{- '<|im_start|>' + message.role + '\n' + content }}
72
+ {%- endif %}
73
+ {%- else %}
74
+ {{- '<|im_start|>' + message.role + '\n' + content }}
75
+ {%- endif %}
76
+ {%- if message.tool_calls %}
77
+ {%- for tool_call in message.tool_calls %}
78
+ {%- if (loop.first and content) or (not loop.first) %}{{- '\n' }}{%- endif %}
79
+ {%- if tool_call.function %}
80
+ {%- set tool_call = tool_call.function %}
81
+ {%- endif %}
82
+ {{- '<tool_call>\n{"name": "' }}
83
+ {{- tool_call.name }}
84
+ {{- '", "arguments": ' }}
85
+ {%- if tool_call.arguments is string %}
86
+ {{- tool_call.arguments }}
87
+ {%- else %}
88
+ {{- tool_call.arguments | tojson }}
89
+ {%- endif %}
90
+ {{- '}\n</tool_call>' }}
91
+ {%- endfor %}
92
+ {%- endif %}
93
+ {{- '<|im_end|>\n' }}
94
+ {%- elif message.role == "tool" %}
95
+ {%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}{{- '<|im_start|>user' }}{%- endif %}
96
+ {{- '\n<tool_response>\n' }}
97
+ {{- content }}
98
+ {{- '\n</tool_response>' }}
99
+ {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}{{- '<|im_end|>\n' }}{%- endif %}
100
+ {%- endif %}
101
+ {%- endfor %}
102
+ {%- if add_generation_prompt %}
103
+ {{- '<|im_start|>assistant\n' }}
104
+ {%- if enable_thinking is defined and enable_thinking is false %}{{- '<think>\n\n</think>\n\n' }}{%- endif %}
105
+ {%- endif %}
config.json ADDED
@@ -0,0 +1,524 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "Qwen3OmniMoeForConditionalGeneration"
4
+ ],
5
+ "assistant_token_id": 77091,
6
+ "code2wav_config": {
7
+ "attention_bias": false,
8
+ "attention_dropout": 0.0,
9
+ "codebook_size": 2048,
10
+ "decoder_dim": 1536,
11
+ "dtype": "bfloat16",
12
+ "hidden_act": "silu",
13
+ "hidden_size": 1024,
14
+ "intermediate_size": 3072,
15
+ "layer_scale_initial_scale": 0.01,
16
+ "max_position_embeddings": 8000,
17
+ "model_type": "",
18
+ "num_attention_heads": 16,
19
+ "num_hidden_layers": 8,
20
+ "num_key_value_heads": 16,
21
+ "num_quantizers": 16,
22
+ "rms_norm_eps": 1e-05,
23
+ "rope_theta": 10000,
24
+ "sliding_window": 72,
25
+ "upsample_rates": [
26
+ 8,
27
+ 5,
28
+ 4,
29
+ 3
30
+ ],
31
+ "upsampling_ratios": [
32
+ 2,
33
+ 2
34
+ ]
35
+ },
36
+ "dtype": "bfloat16",
37
+ "enable_audio_output": false,
38
+ "im_end_token_id": 151645,
39
+ "im_start_token_id": 151644,
40
+ "model_type": "qwen3_omni_moe",
41
+ "system_token_id": 8948,
42
+ "talker_config": {
43
+ "accept_hidden_layer": 18,
44
+ "audio_start_token_id": 151669,
45
+ "audio_token_id": 151646,
46
+ "code_predictor_config": {
47
+ "_name_or_path": "",
48
+ "add_cross_attention": false,
49
+ "architectures": null,
50
+ "attention_bias": false,
51
+ "attention_dropout": 0,
52
+ "bad_words_ids": null,
53
+ "begin_suppress_tokens": null,
54
+ "bos_token_id": null,
55
+ "chunk_size_feed_forward": 0,
56
+ "cross_attention_hidden_size": null,
57
+ "decoder_start_token_id": null,
58
+ "diversity_penalty": 0.0,
59
+ "do_sample": false,
60
+ "dtype": null,
61
+ "early_stopping": false,
62
+ "encoder_no_repeat_ngram_size": 0,
63
+ "eos_token_id": null,
64
+ "exponential_decay_length_penalty": null,
65
+ "finetuning_task": null,
66
+ "forced_bos_token_id": null,
67
+ "forced_eos_token_id": null,
68
+ "head_dim": 128,
69
+ "hidden_act": "silu",
70
+ "hidden_size": 1024,
71
+ "id2label": {
72
+ "0": "LABEL_0",
73
+ "1": "LABEL_1"
74
+ },
75
+ "initializer_range": 0.02,
76
+ "intermediate_size": 3072,
77
+ "is_decoder": false,
78
+ "is_encoder_decoder": false,
79
+ "label2id": {
80
+ "LABEL_0": 0,
81
+ "LABEL_1": 1
82
+ },
83
+ "layer_types": [
84
+ "full_attention",
85
+ "full_attention",
86
+ "full_attention",
87
+ "full_attention",
88
+ "full_attention"
89
+ ],
90
+ "length_penalty": 1.0,
91
+ "max_length": 20,
92
+ "max_position_embeddings": 32768,
93
+ "min_length": 0,
94
+ "model_type": "qwen3_omni_moe_talker_code_predictor",
95
+ "no_repeat_ngram_size": 0,
96
+ "num_attention_heads": 16,
97
+ "num_beam_groups": 1,
98
+ "num_beams": 1,
99
+ "num_code_groups": 32,
100
+ "num_hidden_layers": 5,
101
+ "num_key_value_heads": 8,
102
+ "num_return_sequences": 1,
103
+ "output_attentions": false,
104
+ "output_hidden_states": false,
105
+ "output_scores": false,
106
+ "pad_token_id": null,
107
+ "prefix": null,
108
+ "problem_type": null,
109
+ "pruned_heads": {},
110
+ "remove_invalid_values": false,
111
+ "repetition_penalty": 1.0,
112
+ "return_dict": true,
113
+ "return_dict_in_generate": false,
114
+ "rms_norm_eps": 1e-06,
115
+ "rope_scaling": null,
116
+ "rope_theta": 10000,
117
+ "sep_token_id": null,
118
+ "sliding_window": null,
119
+ "suppress_tokens": null,
120
+ "task_specific_params": null,
121
+ "temperature": 1.0,
122
+ "tf_legacy_loss": false,
123
+ "tie_encoder_decoder": false,
124
+ "tie_word_embeddings": false,
125
+ "tokenizer_class": null,
126
+ "top_k": 50,
127
+ "top_p": 1.0,
128
+ "torchscript": false,
129
+ "typical_p": 1.0,
130
+ "use_bfloat16": false,
131
+ "use_cache": true,
132
+ "vocab_size": 2048
133
+ },
134
+ "codec_bos_id": 4197,
135
+ "codec_eos_token_id": 4198,
136
+ "codec_nothink_id": 4203,
137
+ "codec_pad_id": 4196,
138
+ "codec_think_bos_id": 4204,
139
+ "codec_think_eos_id": 4205,
140
+ "dtype": "bfloat16",
141
+ "image_token_id": 151655,
142
+ "model_type": "",
143
+ "num_code_groups": 32,
144
+ "position_id_per_seconds": 25,
145
+ "speaker_id": null,
146
+ "text_config": {
147
+ "_name_or_path": "",
148
+ "add_cross_attention": false,
149
+ "architectures": null,
150
+ "attention_bias": false,
151
+ "attention_dropout": 0,
152
+ "bad_words_ids": null,
153
+ "begin_suppress_tokens": null,
154
+ "bos_token_id": null,
155
+ "chunk_size_feed_forward": 0,
156
+ "cross_attention_hidden_size": null,
157
+ "decoder_sparse_step": 1,
158
+ "decoder_start_token_id": null,
159
+ "diversity_penalty": 0.0,
160
+ "do_sample": false,
161
+ "dtype": null,
162
+ "early_stopping": false,
163
+ "encoder_no_repeat_ngram_size": 0,
164
+ "eos_token_id": null,
165
+ "exponential_decay_length_penalty": null,
166
+ "finetuning_task": null,
167
+ "forced_bos_token_id": null,
168
+ "forced_eos_token_id": null,
169
+ "hidden_act": "silu",
170
+ "hidden_size": 1024,
171
+ "id2label": {
172
+ "0": "LABEL_0",
173
+ "1": "LABEL_1"
174
+ },
175
+ "initializer_range": 0.02,
176
+ "intermediate_size": 2048,
177
+ "is_decoder": false,
178
+ "is_encoder_decoder": false,
179
+ "label2id": {
180
+ "LABEL_0": 0,
181
+ "LABEL_1": 1
182
+ },
183
+ "length_penalty": 1.0,
184
+ "max_length": 20,
185
+ "max_position_embeddings": 32768,
186
+ "min_length": 0,
187
+ "mlp_only_layers": [],
188
+ "model_type": "qwen3_omni_moe_talker_text",
189
+ "moe_intermediate_size": 384,
190
+ "no_repeat_ngram_size": 0,
191
+ "norm_topk_prob": false,
192
+ "num_attention_heads": 16,
193
+ "num_beam_groups": 1,
194
+ "num_beams": 1,
195
+ "num_experts": 128,
196
+ "num_experts_per_tok": 8,
197
+ "num_hidden_layers": 20,
198
+ "num_key_value_heads": 2,
199
+ "num_return_sequences": 1,
200
+ "output_attentions": false,
201
+ "output_hidden_states": false,
202
+ "output_router_logits": false,
203
+ "output_scores": false,
204
+ "pad_token_id": null,
205
+ "prefix": null,
206
+ "problem_type": null,
207
+ "pruned_heads": {},
208
+ "remove_invalid_values": false,
209
+ "repetition_penalty": 1.0,
210
+ "return_dict": true,
211
+ "return_dict_in_generate": false,
212
+ "rms_norm_eps": 1e-06,
213
+ "rope_scaling": null,
214
+ "rope_theta": 10000,
215
+ "router_aux_loss_coef": 0.001,
216
+ "sep_token_id": null,
217
+ "sliding_window": null,
218
+ "suppress_tokens": null,
219
+ "task_specific_params": null,
220
+ "temperature": 1.0,
221
+ "tf_legacy_loss": false,
222
+ "tie_encoder_decoder": false,
223
+ "tie_word_embeddings": false,
224
+ "tokenizer_class": null,
225
+ "top_k": 50,
226
+ "top_p": 1.0,
227
+ "torchscript": false,
228
+ "typical_p": 1.0,
229
+ "use_bfloat16": false,
230
+ "use_cache": true,
231
+ "vocab_size": 3072
232
+ },
233
+ "thinker_hidden_size": 2048,
234
+ "video_token_id": 151656,
235
+ "vision_start_token_id": 151652
236
+ },
237
+ "thinker_config": {
238
+ "audio_config": {
239
+ "_name_or_path": "",
240
+ "activation_dropout": 0,
241
+ "activation_function": "gelu",
242
+ "add_cross_attention": false,
243
+ "architectures": null,
244
+ "attention_dropout": 0,
245
+ "bad_words_ids": null,
246
+ "begin_suppress_tokens": null,
247
+ "bos_token_id": null,
248
+ "chunk_size_feed_forward": 0,
249
+ "conv_chunksize": 500,
250
+ "cross_attention_hidden_size": null,
251
+ "d_model": 1280,
252
+ "decoder_start_token_id": null,
253
+ "diversity_penalty": 0.0,
254
+ "do_sample": false,
255
+ "downsample_hidden_size": 480,
256
+ "dropout": 0,
257
+ "dtype": null,
258
+ "early_stopping": false,
259
+ "encoder_attention_heads": 20,
260
+ "encoder_ffn_dim": 5120,
261
+ "encoder_layers": 32,
262
+ "encoder_no_repeat_ngram_size": 0,
263
+ "eos_token_id": null,
264
+ "exponential_decay_length_penalty": null,
265
+ "finetuning_task": null,
266
+ "forced_bos_token_id": null,
267
+ "forced_eos_token_id": null,
268
+ "id2label": {
269
+ "0": "LABEL_0",
270
+ "1": "LABEL_1"
271
+ },
272
+ "initializer_range": 0.02,
273
+ "is_decoder": false,
274
+ "is_encoder_decoder": false,
275
+ "label2id": {
276
+ "LABEL_0": 0,
277
+ "LABEL_1": 1
278
+ },
279
+ "length_penalty": 1.0,
280
+ "max_length": 20,
281
+ "max_source_positions": 1500,
282
+ "min_length": 0,
283
+ "model_type": "qwen3_omni_moe_audio_encoder",
284
+ "n_window": 50,
285
+ "n_window_infer": 800,
286
+ "no_repeat_ngram_size": 0,
287
+ "num_beam_groups": 1,
288
+ "num_beams": 1,
289
+ "num_hidden_layers": 32,
290
+ "num_mel_bins": 128,
291
+ "num_return_sequences": 1,
292
+ "output_attentions": false,
293
+ "output_dim": 2048,
294
+ "output_hidden_states": false,
295
+ "output_scores": false,
296
+ "pad_token_id": null,
297
+ "prefix": null,
298
+ "problem_type": null,
299
+ "pruned_heads": {},
300
+ "remove_invalid_values": false,
301
+ "repetition_penalty": 1.0,
302
+ "return_dict": true,
303
+ "return_dict_in_generate": false,
304
+ "scale_embedding": false,
305
+ "sep_token_id": null,
306
+ "suppress_tokens": null,
307
+ "task_specific_params": null,
308
+ "temperature": 1.0,
309
+ "tf_legacy_loss": false,
310
+ "tie_encoder_decoder": false,
311
+ "tie_word_embeddings": true,
312
+ "tokenizer_class": null,
313
+ "top_k": 50,
314
+ "top_p": 1.0,
315
+ "torchscript": false,
316
+ "typical_p": 1.0,
317
+ "use_bfloat16": false
318
+ },
319
+ "audio_end_token_id": 151670,
320
+ "audio_start_token_id": 151669,
321
+ "audio_token_id": 151675,
322
+ "dtype": "bfloat16",
323
+ "image_token_id": 151655,
324
+ "initializer_range": 0.02,
325
+ "model_type": "qwen3_omni_moe_thinker",
326
+ "position_id_per_seconds": 13,
327
+ "seconds_per_chunk": 2,
328
+ "text_config": {
329
+ "_name_or_path": "",
330
+ "add_cross_attention": false,
331
+ "architectures": null,
332
+ "attention_bias": false,
333
+ "attention_dropout": 0.0,
334
+ "bad_words_ids": null,
335
+ "begin_suppress_tokens": null,
336
+ "bos_token_id": null,
337
+ "chunk_size_feed_forward": 0,
338
+ "cross_attention_hidden_size": null,
339
+ "decoder_sparse_step": 1,
340
+ "decoder_start_token_id": null,
341
+ "diversity_penalty": 0.0,
342
+ "do_sample": false,
343
+ "dtype": null,
344
+ "early_stopping": false,
345
+ "encoder_no_repeat_ngram_size": 0,
346
+ "eos_token_id": null,
347
+ "exponential_decay_length_penalty": null,
348
+ "finetuning_task": null,
349
+ "forced_bos_token_id": null,
350
+ "forced_eos_token_id": null,
351
+ "head_dim": 128,
352
+ "hidden_act": "silu",
353
+ "hidden_size": 2048,
354
+ "id2label": {
355
+ "0": "LABEL_0",
356
+ "1": "LABEL_1"
357
+ },
358
+ "initializer_range": 0.02,
359
+ "intermediate_size": 768,
360
+ "is_decoder": false,
361
+ "is_encoder_decoder": false,
362
+ "label2id": {
363
+ "LABEL_0": 0,
364
+ "LABEL_1": 1
365
+ },
366
+ "length_penalty": 1.0,
367
+ "max_length": 20,
368
+ "max_position_embeddings": 65536,
369
+ "min_length": 0,
370
+ "mlp_only_layers": [],
371
+ "model_type": "qwen3_omni_moe_text",
372
+ "moe_intermediate_size": 768,
373
+ "no_repeat_ngram_size": 0,
374
+ "norm_topk_prob": true,
375
+ "num_attention_heads": 32,
376
+ "num_beam_groups": 1,
377
+ "num_beams": 1,
378
+ "num_experts": 128,
379
+ "num_experts_per_tok": 8,
380
+ "num_hidden_layers": 48,
381
+ "num_key_value_heads": 4,
382
+ "num_return_sequences": 1,
383
+ "output_attentions": false,
384
+ "output_hidden_states": false,
385
+ "output_router_logits": false,
386
+ "output_scores": false,
387
+ "pad_token_id": null,
388
+ "prefix": null,
389
+ "problem_type": null,
390
+ "pruned_heads": {},
391
+ "remove_invalid_values": false,
392
+ "repetition_penalty": 1.0,
393
+ "return_dict": true,
394
+ "return_dict_in_generate": false,
395
+ "rms_norm_eps": 1e-06,
396
+ "rope_scaling": {
397
+ "interleaved": true,
398
+ "mrope_interleaved": true,
399
+ "mrope_section": [
400
+ 24,
401
+ 20,
402
+ 20
403
+ ],
404
+ "rope_type": "default",
405
+ "type": "default"
406
+ },
407
+ "rope_theta": 1000000,
408
+ "router_aux_loss_coef": 0.001,
409
+ "sep_token_id": null,
410
+ "shared_expert_intermediate_size": 0,
411
+ "sliding_window": null,
412
+ "suppress_tokens": null,
413
+ "task_specific_params": null,
414
+ "temperature": 1.0,
415
+ "tf_legacy_loss": false,
416
+ "tie_encoder_decoder": false,
417
+ "tie_word_embeddings": false,
418
+ "tokenizer_class": null,
419
+ "top_k": 50,
420
+ "top_p": 1.0,
421
+ "torchscript": false,
422
+ "typical_p": 1.0,
423
+ "use_bfloat16": false,
424
+ "use_cache": true,
425
+ "use_qk_norm": true,
426
+ "use_sliding_window": false,
427
+ "vocab_size": 152064
428
+ },
429
+ "user_token_id": 872,
430
+ "video_token_id": 151656,
431
+ "vision_config": {
432
+ "_name_or_path": "",
433
+ "add_cross_attention": false,
434
+ "apply_vit_abs_pos_embed": true,
435
+ "architectures": null,
436
+ "bad_words_ids": null,
437
+ "begin_suppress_tokens": null,
438
+ "bos_token_id": null,
439
+ "chunk_size_feed_forward": 0,
440
+ "cross_attention_hidden_size": null,
441
+ "decoder_start_token_id": null,
442
+ "deepstack_visual_indexes": [
443
+ 8,
444
+ 16,
445
+ 24
446
+ ],
447
+ "depth": 27,
448
+ "diversity_penalty": 0.0,
449
+ "do_sample": false,
450
+ "dtype": null,
451
+ "early_stopping": false,
452
+ "encoder_no_repeat_ngram_size": 0,
453
+ "eos_token_id": null,
454
+ "exponential_decay_length_penalty": null,
455
+ "finetuning_task": null,
456
+ "forced_bos_token_id": null,
457
+ "forced_eos_token_id": null,
458
+ "hidden_act": "gelu_pytorch_tanh",
459
+ "hidden_size": 1152,
460
+ "id2label": {
461
+ "0": "LABEL_0",
462
+ "1": "LABEL_1"
463
+ },
464
+ "image_size": 768,
465
+ "in_channels": 3,
466
+ "in_chans": 3,
467
+ "initializer_range": 0.02,
468
+ "intermediate_size": 4304,
469
+ "is_decoder": false,
470
+ "is_encoder_decoder": false,
471
+ "label2id": {
472
+ "LABEL_0": 0,
473
+ "LABEL_1": 1
474
+ },
475
+ "length_penalty": 1.0,
476
+ "max_length": 20,
477
+ "min_length": 0,
478
+ "model_type": "qwen3_omni_moe_vision_encoder",
479
+ "no_repeat_ngram_size": 0,
480
+ "num_beam_groups": 1,
481
+ "num_beams": 1,
482
+ "num_heads": 16,
483
+ "num_position_embeddings": 2304,
484
+ "num_return_sequences": 1,
485
+ "out_hidden_size": 2048,
486
+ "output_attentions": false,
487
+ "output_hidden_states": false,
488
+ "output_scores": false,
489
+ "pad_token_id": null,
490
+ "patch_size": 16,
491
+ "prefix": null,
492
+ "problem_type": null,
493
+ "pruned_heads": {},
494
+ "remove_invalid_values": false,
495
+ "repetition_penalty": 1.0,
496
+ "return_dict": true,
497
+ "return_dict_in_generate": false,
498
+ "sep_token_id": null,
499
+ "spatial_merge_size": 2,
500
+ "spatial_patch_size": 16,
501
+ "suppress_tokens": null,
502
+ "task_specific_params": null,
503
+ "temperature": 1.0,
504
+ "temporal_patch_size": 2,
505
+ "tf_legacy_loss": false,
506
+ "tie_encoder_decoder": false,
507
+ "tie_word_embeddings": true,
508
+ "tokenizer_class": null,
509
+ "tokens_per_second": 2,
510
+ "top_k": 50,
511
+ "top_p": 1.0,
512
+ "torchscript": false,
513
+ "typical_p": 1.0,
514
+ "use_bfloat16": false
515
+ },
516
+ "vision_end_token_id": 151653,
517
+ "vision_start_token_id": 151652
518
+ },
519
+ "transformers_version": "4.57.6",
520
+ "tts_bos_token_id": 151672,
521
+ "tts_eos_token_id": 151673,
522
+ "tts_pad_token_id": 151671,
523
+ "user_token_id": 872
524
+ }
generation_config.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "do_sample": true,
3
+ "max_new_tokens": 32768,
4
+ "temperature": 0.6,
5
+ "top_k": 20,
6
+ "top_p": 0.95,
7
+ "transformers_version": "4.57.6"
8
+ }
merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
model-00001-of-00013.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2f91bf7e7ac2c1afd850508e3db645938af0e2763c594116ea18146b74977a3e
3
+ size 4997899632
model-00002-of-00013.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dbe68e3f0f8d6e5064e03fbfe446e8dad233a04a094df3b3261b330c8d395fbe
3
+ size 4997754216
model-00003-of-00013.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0601e1e9c383a10867724dc446e72396ce317383fefe796bfebc6636fbb3481e
3
+ size 4997754216
model-00004-of-00013.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a8d1cecdbe31dbaece53be5d42404133e6408f23774b7feed2a0a90818d14db4
3
+ size 4997755648
model-00005-of-00013.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d01c32ced1c4a7d06915416fc1d5da27ced90a7140bdcee880f18b914313fe60
3
+ size 4997755792
model-00006-of-00013.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1b45cd5dfc1affb1a2618c3ebb8cdbdade0595df55397e012ad9efe75208eb4d
3
+ size 4997755792
model-00007-of-00013.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:94028383a60ce0e732a0ed47c97c1355a83b3e73c1cbe043670eb290657cf0de
3
+ size 4997755792
model-00008-of-00013.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1f508c0b5ca0e823402ef96d51ead216745001eba7d854f1cfd75d32dd50ac47
3
+ size 4997755792
model-00009-of-00013.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2687803c84fd71ebc38547bc95336c2fbeb2a5288573daaeb916f4a5a383fd3c
3
+ size 4997755792
model-00010-of-00013.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3107a7d11caad2a4f4518d2e281cfd89c888971ff3605c40f561403de3543ac9
3
+ size 4997755792
model-00011-of-00013.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f3acc7275c896c4dd539eb5de39ef660e2e3b401c4fe46ec477058e88575204c
3
+ size 4997755792
model-00012-of-00013.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:564c5bccd5e0ebbb3e417442da39e9aae59458a91e7610dafba577b9ddb2c4c1
3
+ size 4997755792
model-00013-of-00013.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e11974655dfc99be61353c0adce879ee63fd9cfc682ccb9e26f990d0bd73eff5
3
+ size 3467789632
model.safetensors.index.json ADDED
The diff for this file is too large to render. See raw diff
 
preprocessor_config.json ADDED
@@ -0,0 +1,31 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "chunk_length": 30,
3
+ "dither": 0.0,
4
+ "feature_extractor_type": "WhisperFeatureExtractor",
5
+ "feature_size": 128,
6
+ "hop_length": 160,
7
+ "image_mean": [
8
+ 0.5,
9
+ 0.5,
10
+ 0.5
11
+ ],
12
+ "image_processor_type": "Qwen2VLImageProcessor",
13
+ "image_std": [
14
+ 0.5,
15
+ 0.5,
16
+ 0.5
17
+ ],
18
+ "max_pixels": 12845056,
19
+ "merge_size": 2,
20
+ "min_pixels": 3136,
21
+ "n_fft": 400,
22
+ "n_samples": 480000,
23
+ "nb_max_frames": 3000,
24
+ "padding_side": "right",
25
+ "padding_value": 0.0,
26
+ "patch_size": 16,
27
+ "processor_class": "Qwen3OmniMoeProcessor",
28
+ "return_attention_mask": true,
29
+ "sampling_rate": 16000,
30
+ "temporal_patch_size": 2
31
+ }
special_tokens_map.json ADDED
@@ -0,0 +1,44 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ "<|im_start|>",
4
+ "<|im_end|>",
5
+ "<|object_ref_start|>",
6
+ "<|object_ref_end|>",
7
+ "<|box_start|>",
8
+ "<|box_end|>",
9
+ "<|quad_start|>",
10
+ "<|quad_end|>",
11
+ "<|vision_start|>",
12
+ "<|vision_end|>",
13
+ "<|vision_pad|>",
14
+ "<|image_pad|>",
15
+ "<|video_pad|>",
16
+ "<|audio_start|>",
17
+ "<|audio_end|>",
18
+ "<tts_pad>",
19
+ "<tts_text_bos>",
20
+ "<tts_text_bos_single>",
21
+ "<|audio_pad|>"
22
+ ],
23
+ "audio_bos_token": "<|audio_start|>",
24
+ "audio_eos_token": "<|audio_end|>",
25
+ "audio_token": "<|audio_pad|>",
26
+ "eos_token": {
27
+ "content": "<|im_end|>",
28
+ "lstrip": false,
29
+ "normalized": false,
30
+ "rstrip": false,
31
+ "single_word": false
32
+ },
33
+ "image_token": "<|image_pad|>",
34
+ "pad_token": {
35
+ "content": "<|endoftext|>",
36
+ "lstrip": false,
37
+ "normalized": false,
38
+ "rstrip": false,
39
+ "single_word": false
40
+ },
41
+ "video_token": "<|video_pad|>",
42
+ "vision_bos_token": "<|vision_start|>",
43
+ "vision_eos_token": "<|vision_end|>"
44
+ }
tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:09267689b8362020b9763b65dd5be7e086b31e28d72e02837a9e781de9a91bc7
3
+ size 11423986
tokenizer_config.json ADDED
@@ -0,0 +1,317 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": false,
3
+ "add_prefix_space": false,
4
+ "added_tokens_decoder": {
5
+ "151643": {
6
+ "content": "<|endoftext|>",
7
+ "lstrip": false,
8
+ "normalized": false,
9
+ "rstrip": false,
10
+ "single_word": false,
11
+ "special": true
12
+ },
13
+ "151644": {
14
+ "content": "<|im_start|>",
15
+ "lstrip": false,
16
+ "normalized": false,
17
+ "rstrip": false,
18
+ "single_word": false,
19
+ "special": true
20
+ },
21
+ "151645": {
22
+ "content": "<|im_end|>",
23
+ "lstrip": false,
24
+ "normalized": false,
25
+ "rstrip": false,
26
+ "single_word": false,
27
+ "special": true
28
+ },
29
+ "151646": {
30
+ "content": "<|object_ref_start|>",
31
+ "lstrip": false,
32
+ "normalized": false,
33
+ "rstrip": false,
34
+ "single_word": false,
35
+ "special": true
36
+ },
37
+ "151647": {
38
+ "content": "<|object_ref_end|>",
39
+ "lstrip": false,
40
+ "normalized": false,
41
+ "rstrip": false,
42
+ "single_word": false,
43
+ "special": true
44
+ },
45
+ "151648": {
46
+ "content": "<|box_start|>",
47
+ "lstrip": false,
48
+ "normalized": false,
49
+ "rstrip": false,
50
+ "single_word": false,
51
+ "special": true
52
+ },
53
+ "151649": {
54
+ "content": "<|box_end|>",
55
+ "lstrip": false,
56
+ "normalized": false,
57
+ "rstrip": false,
58
+ "single_word": false,
59
+ "special": true
60
+ },
61
+ "151650": {
62
+ "content": "<|quad_start|>",
63
+ "lstrip": false,
64
+ "normalized": false,
65
+ "rstrip": false,
66
+ "single_word": false,
67
+ "special": true
68
+ },
69
+ "151651": {
70
+ "content": "<|quad_end|>",
71
+ "lstrip": false,
72
+ "normalized": false,
73
+ "rstrip": false,
74
+ "single_word": false,
75
+ "special": true
76
+ },
77
+ "151652": {
78
+ "content": "<|vision_start|>",
79
+ "lstrip": false,
80
+ "normalized": false,
81
+ "rstrip": false,
82
+ "single_word": false,
83
+ "special": true
84
+ },
85
+ "151653": {
86
+ "content": "<|vision_end|>",
87
+ "lstrip": false,
88
+ "normalized": false,
89
+ "rstrip": false,
90
+ "single_word": false,
91
+ "special": true
92
+ },
93
+ "151654": {
94
+ "content": "<|vision_pad|>",
95
+ "lstrip": false,
96
+ "normalized": false,
97
+ "rstrip": false,
98
+ "single_word": false,
99
+ "special": true
100
+ },
101
+ "151655": {
102
+ "content": "<|image_pad|>",
103
+ "lstrip": false,
104
+ "normalized": false,
105
+ "rstrip": false,
106
+ "single_word": false,
107
+ "special": true
108
+ },
109
+ "151656": {
110
+ "content": "<|video_pad|>",
111
+ "lstrip": false,
112
+ "normalized": false,
113
+ "rstrip": false,
114
+ "single_word": false,
115
+ "special": true
116
+ },
117
+ "151657": {
118
+ "content": "<tool_call>",
119
+ "lstrip": false,
120
+ "normalized": false,
121
+ "rstrip": false,
122
+ "single_word": false,
123
+ "special": false
124
+ },
125
+ "151658": {
126
+ "content": "</tool_call>",
127
+ "lstrip": false,
128
+ "normalized": false,
129
+ "rstrip": false,
130
+ "single_word": false,
131
+ "special": false
132
+ },
133
+ "151659": {
134
+ "content": "<|fim_prefix|>",
135
+ "lstrip": false,
136
+ "normalized": false,
137
+ "rstrip": false,
138
+ "single_word": false,
139
+ "special": false
140
+ },
141
+ "151660": {
142
+ "content": "<|fim_middle|>",
143
+ "lstrip": false,
144
+ "normalized": false,
145
+ "rstrip": false,
146
+ "single_word": false,
147
+ "special": false
148
+ },
149
+ "151661": {
150
+ "content": "<|fim_suffix|>",
151
+ "lstrip": false,
152
+ "normalized": false,
153
+ "rstrip": false,
154
+ "single_word": false,
155
+ "special": false
156
+ },
157
+ "151662": {
158
+ "content": "<|fim_pad|>",
159
+ "lstrip": false,
160
+ "normalized": false,
161
+ "rstrip": false,
162
+ "single_word": false,
163
+ "special": false
164
+ },
165
+ "151663": {
166
+ "content": "<|repo_name|>",
167
+ "lstrip": false,
168
+ "normalized": false,
169
+ "rstrip": false,
170
+ "single_word": false,
171
+ "special": false
172
+ },
173
+ "151664": {
174
+ "content": "<|file_sep|>",
175
+ "lstrip": false,
176
+ "normalized": false,
177
+ "rstrip": false,
178
+ "single_word": false,
179
+ "special": false
180
+ },
181
+ "151665": {
182
+ "content": "<tool_response>",
183
+ "lstrip": false,
184
+ "normalized": false,
185
+ "rstrip": false,
186
+ "single_word": false,
187
+ "special": false
188
+ },
189
+ "151666": {
190
+ "content": "</tool_response>",
191
+ "lstrip": false,
192
+ "normalized": false,
193
+ "rstrip": false,
194
+ "single_word": false,
195
+ "special": false
196
+ },
197
+ "151667": {
198
+ "content": "<think>",
199
+ "lstrip": false,
200
+ "normalized": false,
201
+ "rstrip": false,
202
+ "single_word": false,
203
+ "special": false
204
+ },
205
+ "151668": {
206
+ "content": "</think>",
207
+ "lstrip": false,
208
+ "normalized": false,
209
+ "rstrip": false,
210
+ "single_word": false,
211
+ "special": false
212
+ },
213
+ "151669": {
214
+ "content": "<|audio_start|>",
215
+ "lstrip": false,
216
+ "normalized": false,
217
+ "rstrip": false,
218
+ "single_word": false,
219
+ "special": true
220
+ },
221
+ "151670": {
222
+ "content": "<|audio_end|>",
223
+ "lstrip": false,
224
+ "normalized": false,
225
+ "rstrip": false,
226
+ "single_word": false,
227
+ "special": true
228
+ },
229
+ "151671": {
230
+ "content": "<tts_pad>",
231
+ "lstrip": false,
232
+ "normalized": false,
233
+ "rstrip": false,
234
+ "single_word": false,
235
+ "special": true
236
+ },
237
+ "151672": {
238
+ "content": "<tts_text_bos>",
239
+ "lstrip": false,
240
+ "normalized": false,
241
+ "rstrip": false,
242
+ "single_word": false,
243
+ "special": true
244
+ },
245
+ "151673": {
246
+ "content": "<tts_text_eod>",
247
+ "lstrip": false,
248
+ "normalized": false,
249
+ "rstrip": false,
250
+ "single_word": false,
251
+ "special": true
252
+ },
253
+ "151674": {
254
+ "content": "<tts_text_bos_single>",
255
+ "lstrip": false,
256
+ "normalized": false,
257
+ "rstrip": false,
258
+ "single_word": false,
259
+ "special": true
260
+ },
261
+ "151675": {
262
+ "content": "<|audio_pad|>",
263
+ "lstrip": false,
264
+ "normalized": false,
265
+ "rstrip": false,
266
+ "single_word": false,
267
+ "special": true
268
+ }
269
+ },
270
+ "additional_special_tokens": [
271
+ "<|im_start|>",
272
+ "<|im_end|>",
273
+ "<|object_ref_start|>",
274
+ "<|object_ref_end|>",
275
+ "<|box_start|>",
276
+ "<|box_end|>",
277
+ "<|quad_start|>",
278
+ "<|quad_end|>",
279
+ "<|vision_start|>",
280
+ "<|vision_end|>",
281
+ "<|vision_pad|>",
282
+ "<|image_pad|>",
283
+ "<|video_pad|>",
284
+ "<|audio_start|>",
285
+ "<|audio_end|>",
286
+ "<tts_pad>",
287
+ "<tts_text_bos>",
288
+ "<tts_text_bos_single>",
289
+ "<|audio_pad|>"
290
+ ],
291
+ "audio_bos_token": "<|audio_start|>",
292
+ "audio_eos_token": "<|audio_end|>",
293
+ "audio_token": "<|audio_pad|>",
294
+ "bos_token": null,
295
+ "clean_up_tokenization_spaces": false,
296
+ "eos_token": "<|im_end|>",
297
+ "errors": "replace",
298
+ "extra_special_tokens": {
299
+ "audio_bos_token": "<|audio_start|>",
300
+ "audio_eos_token": "<|audio_end|>",
301
+ "audio_token": "<|audio_pad|>",
302
+ "image_token": "<|image_pad|>",
303
+ "video_token": "<|video_pad|>",
304
+ "vision_bos_token": "<|vision_start|>",
305
+ "vision_eos_token": "<|vision_end|>"
306
+ },
307
+ "image_token": "<|image_pad|>",
308
+ "model_max_length": 131072,
309
+ "pad_token": "<|endoftext|>",
310
+ "processor_class": "Qwen3OmniMoeProcessor",
311
+ "split_special_tokens": false,
312
+ "tokenizer_class": "Qwen2Tokenizer",
313
+ "unk_token": null,
314
+ "video_token": "<|video_pad|>",
315
+ "vision_bos_token": "<|vision_start|>",
316
+ "vision_eos_token": "<|vision_end|>"
317
+ }
video_preprocessor_config.json ADDED
@@ -0,0 +1,54 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "crop_size": null,
3
+ "data_format": "channels_first",
4
+ "default_to_square": true,
5
+ "device": null,
6
+ "dither": 0.0,
7
+ "do_center_crop": null,
8
+ "do_convert_rgb": true,
9
+ "do_normalize": true,
10
+ "do_rescale": true,
11
+ "do_resize": true,
12
+ "do_sample_frames": false,
13
+ "feature_extractor_type": "WhisperFeatureExtractor",
14
+ "feature_size": 128,
15
+ "fps": null,
16
+ "hop_length": 160,
17
+ "image_mean": [
18
+ 0.5,
19
+ 0.5,
20
+ 0.5
21
+ ],
22
+ "image_std": [
23
+ 0.5,
24
+ 0.5,
25
+ 0.5
26
+ ],
27
+ "input_data_format": null,
28
+ "max_frames": 768,
29
+ "max_pixels": 12845056,
30
+ "merge_size": 2,
31
+ "min_frames": 4,
32
+ "min_pixels": 3136,
33
+ "n_fft": 400,
34
+ "n_samples": 4800000,
35
+ "nb_max_frames": 30000,
36
+ "num_frames": null,
37
+ "pad_size": null,
38
+ "padding_side": "right",
39
+ "padding_value": 0.0,
40
+ "patch_size": 16,
41
+ "processor_class": "Qwen3OmniMoeProcessor",
42
+ "resample": 3,
43
+ "rescale_factor": 0.00392156862745098,
44
+ "return_attention_mask": true,
45
+ "return_metadata": false,
46
+ "sampling_rate": 16000,
47
+ "size": {
48
+ "longest_edge": 12845056,
49
+ "shortest_edge": 3136
50
+ },
51
+ "temporal_patch_size": 2,
52
+ "video_metadata": null,
53
+ "video_processor_type": "Qwen2VLVideoProcessor"
54
+ }
vocab.json ADDED
The diff for this file is too large to render. See raw diff