{ "aggregator_config": { "attn_mask_enabled": false, "depth": 8, "dropout": 0.0, "hidden_size": 1024, "mlp_ratio": 4, "num_heads": 16 }, "architectures": [ "BailingMMNativeForConditionalGeneration" ], "audio_tokenizer_config": { "architectures": [ "AudioVAE" ], "dec_kwargs": { "backbone": { "_attn_implementation": "flash_attention_2", "attention_dropout": 0.0, "attn_implementation": null, "bos_token_id": 151643, "eos_token_id": 151645, "hidden_act": "silu", "hidden_size": 896, "initializer_range": 0.02, "intermediate_size": 4864, "is_causal": true, "max_position_embeddings": 32768, "max_window_layers": 0, "model_type": "qwen2", "num_attention_heads": 14, "num_hidden_layers": 24, "num_key_value_heads": 2, "rms_norm_eps": 1e-06, "rope_theta": 1000000.0, "sliding_window": 64, "tie_word_embeddings": true, "torch_dtype": "bfloat16", "transformers_version": "4.43.1", "use_cache": false, "use_sliding_window": true, "vocab_size": 1 }, "latent_dim": 64, "output_dim": 882 }, "enc_kwargs": { "backbone": { "_attn_implementation": "flash_attention_2", "attention_dropout": 0.0, "attn_implementation": null, "bos_token_id": 151643, "eos_token_id": 151645, "hidden_act": "silu", "hidden_size": 896, "initializer_range": 0.02, "intermediate_size": 4864, "is_causal": true, "max_position_embeddings": 32768, "max_window_layers": 0, "model_type": "qwen2", "num_attention_heads": 14, "num_hidden_layers": 24, "num_key_value_heads": 2, "rms_norm_eps": 1e-06, "rope_theta": 1000000.0, "sliding_window": 64, "tie_word_embeddings": true, "torch_dtype": "bfloat16", "transformers_version": "4.43.1", "use_cache": false, "use_sliding_window": true, "vocab_size": 1 }, "hop_size": 882, "input_dim": 882, "latent_dim": 64 }, "hifi_gan_disc_kwargs": { "channel_increasing_factor": 4, "channels": 16, "max_downsample_channels": 512, "periods": [ 2, 3, 5, 7, 11 ] }, "init_method": "kaiming", "lambda_adv": 1.0, "lambda_disc": 1.0, "lambda_feat_match_loss": 1.0, "lambda_mel_loss": 1.0, "lambda_semantic": 2.0, "patch_size": 4, "sample_rate": 44100, "semantic_module_kwargs": null, "spec_disc_kwargs": { "channels": 32, "downsample_scales": [ 2, 2, 2 ], "in_channels": 1, "kernel_sizes": [ 5, 3 ], "max_downsample_channels": 512, "out_channels": 1, "stft_params": { "fft_sizes": [ 78, 126, 206, 334, 542, 876, 1418, 2296 ], "hop_sizes": [ 39, 63, 103, 167, 271, 438, 709, 1148 ], "win_lengths": [ 78, 126, 206, 334, 542, 876, 1418, 2296 ], "window": "hann_window" }, "use_weight_norm": true }, "torch_dtype": "bfloat16", "transformers_version": "4.46.3" }, "auto_map": { "AutoConfig": "configuration_bailingmm.BailingMMConfig" }, "ditar_config": { "attn_backend": "torch", "attn_mask_enabled": false, "depth": 8, "dropout": 0.0, "hidden_size": 1024, "history_patch_size": 1, "mlp_ratio": 4, "num_heads": 16, "patch_size": 1 }, "llm_config": { "architectures": [ "Qwen2ForCausalLM" ], "attention_dropout": 0.0, "bos_token_id": 151643, "eos_token_id": 151645, "hidden_act": "silu", "hidden_size": 896, "initializer_range": 0.02, "intermediate_size": 4864, "max_position_embeddings": 32768, "max_window_layers": 21, "model_type": "qwen2", "num_attention_heads": 14, "num_hidden_layers": 24, "num_key_value_heads": 2, "rms_norm_eps": 1e-06, "rope_theta": 1000000.0, "sliding_window": 32768, "tie_word_embeddings": true, "torch_dtype": "bfloat16", "transformers_version": "4.43.1", "use_cache": true, "use_sliding_window": false, "vocab_size": 151936 }, "mlp_depth": 2, "model_type": "dense", "torch_dtype": "bfloat16", "transformers_version": "4.52.4", "callme": { "converted_from": "inclusionAI/Ming-3.0-tts-0.5B", "dropped_prefixes": [ "audio.encoder.", "model.lm_head." ], "weight_names": "upstream (unchanged)", "quantization": { "bits": 8, "group_size": 64, "mode": "affine", "scope": [ "model.model.layers.", "linear_proj_audio.blocks.", "flowloss.cfm.model.blocks." ] } } }