config.json

{
  "adapter_hidden_size": 8192,
  "architectures": [
    "MusciForCausalLM"
  ],
  "attention_bias": false,
  "attention_dropout": 0.0,
  "audio_config": {
    "activation_dropout": 0,
    "activation_function": "gelu",
    "attention_dropout": 0,
    "conv_chunksize": 500,
    "d_model": 1280,
    "downsample_hidden_size": 480,
    "dropout": 0,
    "encoder_attention_heads": 20,
    "encoder_ffn_dim": 5120,
    "encoder_layers": 32,
    "initializer_range": 0.02,
    "max_source_positions": 1500,
    "model_type": "qwen3_omni_moe_audio_encoder",
    "n_window": 50,
    "n_window_infer": 800,
    "num_hidden_layers": 32,
    "num_mel_bins": 128,
    "output_dim": 2048,
    "scale_embedding": false
  },
  "dtype": "bfloat16",
  "head_dim": 128,
  "hidden_act": "silu",
  "hidden_size": 4096,
  "ignore_index": -100,
  "initializer_range": 0.02,
  "intermediate_size": 22016,
  "language_config": {
    "architectures": [
      "Qwen3ForCausalLM"
    ],
    "attention_bias": false,
    "attention_dropout": 0.0,
    "bos_token_id": 151643,
    "dtype": "bfloat16",
    "eos_token_id": 151645,
    "head_dim": 128,
    "hidden_act": "silu",
    "hidden_size": 2048,
    "initializer_range": 0.02,
    "intermediate_size": 6144,
    "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",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention"
    ],
    "max_position_embeddings": 40960,
    "max_window_layers": 28,
    "model_type": "qwen3",
    "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,
    "tie_word_embeddings": true,
    "use_cache": true,
    "use_sliding_window": false,
    "vocab_size": 151936
  },
  "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",
    "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": "musci",
  "num_attention_heads": 32,
  "num_hidden_layers": 28,
  "num_key_value_heads": 32,
  "rms_norm_eps": 1e-06,
  "rope_scaling": null,
  "rope_theta": 10000.0,
  "sliding_window": null,
  "tie_word_embeddings": false,
  "transformers_version": "4.57.1",
  "use_cache": true,
  "use_sliding_window": false,
  "vocab_size": 151936,
  "auto_map": {
    "AutoConfig": "modeling_Musci.MusciConfig",
    "AutoModelForCausalLM": "modeling_Musci.MusciForCausalLM"
  }
}