{ "architectures": [ "MGMTTSForCausalLM" ], "attention_bias": false, "attention_dropout": 0.0, "bos_token_id": 151643, "eos_token_id": 151645, "head_dim": 128, "hidden_act": "silu", "hidden_size": 2048, "initializer_range": 0.02, "intermediate_size": 6144, "max_position_embeddings": 40960, "max_window_layers": 28, "mm_use_speech_start_end": true, "model_type": "MGMTTS", "num_attention_heads": 16, "num_hidden_layers": 28, "num_key_value_heads": 8, "rms_norm_eps": 1e-06, "rope_scaling": null, "rope_theta": 1000000, "sliding_window": null, "speech_token_count": 4, "speech_token_pre_pad": 4, "speech_token_type": "cosyvoice", "tie_word_embeddings": true, "tokenizer_speech_size": 6563, "tokenizer_text_size": 151672, "torch_dtype": "bfloat16", "training_tts": true, "transformers_version": "4.52.3", "tts_adapter_config": "[6,2048,16_2,8192]", "tts_adapter_type": "qwen3", "tune_tts_adapter_only": false, "use_cache": true, "use_sliding_window": false, "vocab_size": 177924 }