| { |
| "architectures": [ |
| "Qwen2_5OmniForConditionalGeneration" |
| ], |
| "dtype": "bfloat16", |
| "enable_audio_output": false, |
| "enable_talker": false, |
| "hidden_size": 896, |
| "intermediate_size": 2432, |
| "model_type": "qwen2_5_omni", |
| "num_attention_heads": 14, |
| "num_hidden_layers": 18, |
| "num_key_value_heads": 2, |
| "talker_config": {}, |
| "thinker_config": { |
| "_attn_implementation_autoset": true, |
| "architectures": [ |
| "Qwen2OmniNaViTThinkerForConditionalGeneration" |
| ], |
| "audio_config": { |
| "_attn_implementation_autoset": true, |
| "_name_or_path": "", |
| "activation_dropout": 0.0, |
| "activation_function": "gelu", |
| "add_cross_attention": false, |
| "architectures": null, |
| "attention_dropout": 0.0, |
| "bos_token_id": null, |
| "chunk_size_feed_forward": 0, |
| "cross_attention_hidden_size": null, |
| "d_model": 256, |
| "decoder_start_token_id": null, |
| "dropout": 0.0, |
| "dtype": null, |
| "encoder_attention_heads": 4, |
| "encoder_ffn_dim": 1024, |
| "encoder_layerdrop": 0.0, |
| "encoder_layers": 8, |
| "eos_token_id": null, |
| "finetuning_task": null, |
| "id2label": { |
| "0": "LABEL_0", |
| "1": "LABEL_1" |
| }, |
| "init_std": 0.02, |
| "initializer_range": 0.02, |
| "is_decoder": false, |
| "is_encoder_decoder": false, |
| "label2id": { |
| "LABEL_0": "LABEL_0", |
| "LABEL_1": "LABEL_1" |
| }, |
| "max_source_positions": 1500, |
| "model_type": "qwen2_5_omni_audio_encoder", |
| "n_window": 100, |
| "num_mel_bins": 128, |
| "output_attentions": false, |
| "output_dim": 896, |
| "output_hidden_states": false, |
| "pad_token_id": null, |
| "prefix": null, |
| "problem_type": null, |
| "pruned_heads": {}, |
| "return_dict": true, |
| "scale_embedding": false, |
| "sep_token_id": null, |
| "task_specific_params": null, |
| "tf_legacy_loss": false, |
| "tie_encoder_decoder": false, |
| "tie_word_embeddings": true, |
| "tokenizer_class": null, |
| "torchscript": false, |
| "use_bfloat16": false |
| }, |
| "audio_end_token_id": 151648, |
| "audio_start_token_id": 151647, |
| "audio_token_index": 151646, |
| "bos_token_id": 151644, |
| "dtype": "bfloat16", |
| "eos_token_id": 151645, |
| "ignore_index": -100, |
| "image_token_index": 151655, |
| "init_std": 0.02, |
| "initializer_range": 0.02, |
| "model_type": "qwen2_5_omni_thinker", |
| "pad_token_id": 151643, |
| "position_id_per_seconds": 25, |
| "seconds_per_chunk": 2, |
| "text_config": { |
| "_attn_implementation_autoset": false, |
| "_name_or_path": "", |
| "add_cross_attention": false, |
| "architectures": null, |
| "attention_dropout": 0.0, |
| "bos_token_id": null, |
| "chunk_size_feed_forward": 0, |
| "cross_attention_hidden_size": null, |
| "decoder_start_token_id": null, |
| "dtype": null, |
| "eos_token_id": null, |
| "finetuning_task": null, |
| "hidden_act": "silu", |
| "hidden_size": 896, |
| "id2label": { |
| "0": "LABEL_0", |
| "1": "LABEL_1" |
| }, |
| "init_std": 0.02, |
| "initializer_range": 0.02, |
| "intermediate_size": 2432, |
| "is_decoder": false, |
| "is_encoder_decoder": false, |
| "label2id": { |
| "LABEL_0": "LABEL_0", |
| "LABEL_1": "LABEL_1" |
| }, |
| "layer_types": [ |
| "full_attention", |
| "full_attention", |
| "full_attention", |
| "full_attention", |
| "full_attention", |
| "full_attention", |
| "full_attention", |
| "full_attention", |
| "full_attention", |
| "full_attention", |
| "full_attention", |
| "full_attention", |
| "full_attention", |
| "full_attention", |
| "full_attention", |
| "full_attention", |
| "full_attention", |
| "full_attention" |
| ], |
| "max_position_embeddings": 32768, |
| "max_window_layers": 28, |
| "model_type": "qwen2_5_omni_text", |
| "mrope_section": [ |
| 10, |
| 10, |
| 20 |
| ], |
| "num_attention_heads": 14, |
| "num_hidden_layers": 18, |
| "num_key_value_heads": 2, |
| "output_attentions": false, |
| "output_hidden_states": false, |
| "pad_token_id": null, |
| "prefix": null, |
| "problem_type": null, |
| "pruned_heads": {}, |
| "return_dict": true, |
| "rms_norm_eps": 1e-06, |
| "rope_parameters": { |
| "mrope_section": [ |
| 10, |
| 10, |
| 20 |
| ], |
| "rope_theta": 1000000.0, |
| "rope_type": "default" |
| }, |
| "sep_token_id": null, |
| "sliding_window": null, |
| "task_specific_params": null, |
| "tf_legacy_loss": false, |
| "tie_encoder_decoder": false, |
| "tie_word_embeddings": false, |
| "tokenizer_class": null, |
| "torchscript": false, |
| "use_bfloat16": false, |
| "use_cache": true, |
| "use_sliding_window": false, |
| "vocab_size": 32000 |
| }, |
| "tie_word_embeddings": false, |
| "user_token_id": 872, |
| "video_token_index": 151656, |
| "vision_config": { |
| "_attn_implementation_autoset": true, |
| "_name_or_path": "", |
| "add_cross_attention": false, |
| "architectures": null, |
| "bos_token_id": null, |
| "chunk_size_feed_forward": 0, |
| "cross_attention_hidden_size": null, |
| "decoder_start_token_id": null, |
| "depth": 12, |
| "dtype": null, |
| "embed_dim": 512, |
| "eos_token_id": null, |
| "finetuning_task": null, |
| "fullatt_block_indexes": [ |
| 3, |
| 7, |
| 11 |
| ], |
| "hidden_act": "silu", |
| "hidden_size": 512, |
| "id2label": { |
| "0": "LABEL_0", |
| "1": "LABEL_1" |
| }, |
| "in_channels": 3, |
| "in_chans": 3, |
| "init_std": 0.02, |
| "initializer_range": 0.02, |
| "intermediate_size": 2048, |
| "is_decoder": false, |
| "is_encoder_decoder": false, |
| "label2id": { |
| "LABEL_0": "LABEL_0", |
| "LABEL_1": "LABEL_1" |
| }, |
| "model_type": "qwen2_5_omni_vision_encoder", |
| "num_heads": 8, |
| "out_hidden_size": 896, |
| "output_attentions": false, |
| "output_hidden_states": false, |
| "pad_token_id": null, |
| "patch_size": 14, |
| "prefix": null, |
| "problem_type": null, |
| "pruned_heads": {}, |
| "return_dict": true, |
| "sep_token_id": null, |
| "spatial_merge_size": 2, |
| "spatial_patch_size": 14, |
| "task_specific_params": null, |
| "temporal_patch_size": 2, |
| "tf_legacy_loss": false, |
| "tie_encoder_decoder": false, |
| "tie_word_embeddings": true, |
| "tokenizer_class": null, |
| "tokens_per_second": 25, |
| "torchscript": false, |
| "use_bfloat16": false, |
| "window_size": 16 |
| }, |
| "vision_end_token_id": 151653, |
| "vision_start_token_id": 151652, |
| "vision_token_id": 151654 |
| }, |
| "token2wav_config": {}, |
| "transformers_version": "5.5.4" |
| } |
