{ "audio_config": { "num_mel_bins": 128, "encoder_layers": 18, "encoder_attention_heads": 14, "encoder_ffn_dim": 3584, "d_model": 896, "dropout": 0, "attention_dropout": 0, "activation_function": "gelu", "activation_dropout": 0, "scale_embedding": false, "initializer_range": 0.02, "max_source_positions": 1500, "n_window": 50, "output_dim": 1024, "n_window_infer": 800, "conv_chunksize": 500, "downsample_hidden_size": 480 }, "text_config": { "model_type": "qwen3", "vocab_size": 151936, "hidden_size": 1024, "intermediate_size": 3072, "num_hidden_layers": 28, "num_attention_heads": 16, "num_key_value_heads": 8, "head_dim": 128, "hidden_act": "silu", "max_position_embeddings": 65536, "initializer_range": 0.02, "rms_norm_eps": 1e-06, "use_cache": true, "tie_word_embeddings": true, "rope_theta": 1000000, "rope_scaling": { "interleaved": true, "mrope_interleaved": true, "mrope_section": [ 24, 20, 20 ], "rope_type": "default", "type": "default" }, "attention_bias": false, "attention_dropout": 0.0 }, "model_type": "qwen3_asr", "model_repo": "mlx-community/Qwen3-ASR-0.6B-5bit", "audio_token_id": 151676, "audio_start_token_id": 151669, "audio_end_token_id": 151670, "support_languages": [ "Chinese", "English", "Cantonese", "Arabic", "German", "French", "Spanish", "Portuguese", "Indonesian", "Italian", "Korean", "Russian", "Thai", "Vietnamese", "Japanese", "Turkish", "Hindi", "Malay", "Dutch", "Swedish", "Danish", "Finnish", "Polish", "Czech", "Filipino", "Persian", "Greek", "Romanian", "Hungarian", "Macedonian" ], "architectures": [ "Qwen3ASRForConditionalGeneration" ] }