| { |
| "architectures": [ |
| "Qwen3OmniMoeThinkerForConditionalGeneration" |
| ], |
| "auto_map": { |
| "AutoConfig": "configuration_qwen3_omni_moe_thinker.Qwen3OmniMoeThinkerConfig", |
| "AutoModelForCausalLM": "modeling_qwen3_omni_moe_thinker.Qwen3OmniMoeThinkerForConditionalGeneration" |
| }, |
| "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 |
| }, |
| "audio_end_token_id": 151670, |
| "audio_start_token_id": 151669, |
| "audio_token_id": 151675, |
| "dtype": "bfloat16", |
| "image_token_id": 151655, |
| "initializer_range": 0.02, |
| "model_type": "qwen3_omni_moe_thinker", |
| "position_id_per_seconds": 13, |
| "quantization_config": { |
| "config_groups": { |
| "group_0": { |
| "format": "pack-quantized", |
| "input_activations": null, |
| "output_activations": null, |
| "targets": [ |
| "Linear" |
| ], |
| "weights": { |
| "actorder": "static", |
| "block_structure": null, |
| "dynamic": false, |
| "group_size": 128, |
| "num_bits": 4, |
| "observer": "minmax", |
| "observer_kwargs": {}, |
| "scale_dtype": null, |
| "strategy": "group", |
| "symmetric": true, |
| "type": "int", |
| "zp_dtype": null |
| } |
| } |
| }, |
| "format": "pack-quantized", |
| "global_compression_ratio": null, |
| "ignore": [ |
| "audio_tower.layers.0.self_attn.k_proj", |
| "audio_tower.layers.0.self_attn.v_proj", |
| "audio_tower.layers.0.self_attn.q_proj", |
| "audio_tower.layers.0.self_attn.out_proj", |
| "audio_tower.layers.0.fc1", |
| "audio_tower.layers.0.fc2", |
| "audio_tower.layers.1.self_attn.k_proj", |
| "audio_tower.layers.1.self_attn.v_proj", |
| "audio_tower.layers.1.self_attn.q_proj", |
| "audio_tower.layers.1.self_attn.out_proj", |
| "audio_tower.layers.1.fc1", |
| "audio_tower.layers.1.fc2", |
| "audio_tower.layers.2.self_attn.k_proj", |
| "audio_tower.layers.2.self_attn.v_proj", |
| "audio_tower.layers.2.self_attn.q_proj", |
| "audio_tower.layers.2.self_attn.out_proj", |
| "audio_tower.layers.2.fc1", |
| "audio_tower.layers.2.fc2", |
| "audio_tower.layers.3.self_attn.k_proj", |
| "audio_tower.layers.3.self_attn.v_proj", |
| "audio_tower.layers.3.self_attn.q_proj", |
| "audio_tower.layers.3.self_attn.out_proj", |
| "audio_tower.layers.3.fc1", |
| "audio_tower.layers.3.fc2", |
| "audio_tower.layers.4.self_attn.k_proj", |
| "audio_tower.layers.4.self_attn.v_proj", |
| "audio_tower.layers.4.self_attn.q_proj", |
| "audio_tower.layers.4.self_attn.out_proj", |
| "audio_tower.layers.4.fc1", |
| "audio_tower.layers.4.fc2", |
| "audio_tower.layers.5.self_attn.k_proj", |
| "audio_tower.layers.5.self_attn.v_proj", |
| "audio_tower.layers.5.self_attn.q_proj", |
| "audio_tower.layers.5.self_attn.out_proj", |
| "audio_tower.layers.5.fc1", |
| "audio_tower.layers.5.fc2", |
| "audio_tower.layers.6.self_attn.k_proj", |
| "audio_tower.layers.6.self_attn.v_proj", |
| "audio_tower.layers.6.self_attn.q_proj", |
| "audio_tower.layers.6.self_attn.out_proj", |
| "audio_tower.layers.6.fc1", |
| "audio_tower.layers.6.fc2", |
| "audio_tower.layers.7.self_attn.k_proj", |
| "audio_tower.layers.7.self_attn.v_proj", |
| "audio_tower.layers.7.self_attn.q_proj", |
| "audio_tower.layers.7.self_attn.out_proj", |
| "audio_tower.layers.7.fc1", |
| "audio_tower.layers.7.fc2", |
| "audio_tower.layers.8.self_attn.k_proj", |
| "audio_tower.layers.8.self_attn.v_proj", |
| "audio_tower.layers.8.self_attn.q_proj", |
| "audio_tower.layers.8.self_attn.out_proj", |
| "audio_tower.layers.8.fc1", |
| "audio_tower.layers.8.fc2", |
| "audio_tower.layers.9.self_attn.k_proj", |
| "audio_tower.layers.9.self_attn.v_proj", |
| "audio_tower.layers.9.self_attn.q_proj", |
| "audio_tower.layers.9.self_attn.out_proj", |
| "audio_tower.layers.9.fc1", |
| "audio_tower.layers.9.fc2", |
| "audio_tower.layers.10.self_attn.k_proj", |
| "audio_tower.layers.10.self_attn.v_proj", |
| "audio_tower.layers.10.self_attn.q_proj", |
| "audio_tower.layers.10.self_attn.out_proj", |
| "audio_tower.layers.10.fc1", |
| "audio_tower.layers.10.fc2", |
| "audio_tower.layers.11.self_attn.k_proj", |
| "audio_tower.layers.11.self_attn.v_proj", |
| "audio_tower.layers.11.self_attn.q_proj", |
| "audio_tower.layers.11.self_attn.out_proj", |
| "audio_tower.layers.11.fc1", |
| "audio_tower.layers.11.fc2", |
| "audio_tower.layers.12.self_attn.k_proj", |
| "audio_tower.layers.12.self_attn.v_proj", |
| "audio_tower.layers.12.self_attn.q_proj", |
| "audio_tower.layers.12.self_attn.out_proj", |
| "audio_tower.layers.12.fc1", |
| "audio_tower.layers.12.fc2", |
| "audio_tower.layers.13.self_attn.k_proj", |
| "audio_tower.layers.13.self_attn.v_proj", |
| "audio_tower.layers.13.self_attn.q_proj", |
| "audio_tower.layers.13.self_attn.out_proj", |
| "audio_tower.layers.13.fc1", |
| "audio_tower.layers.13.fc2", |
| "audio_tower.layers.14.self_attn.k_proj", |
| "audio_tower.layers.14.self_attn.v_proj", |
| "audio_tower.layers.14.self_attn.q_proj", |
| "audio_tower.layers.14.self_attn.out_proj", |
| "audio_tower.layers.14.fc1", |
| "audio_tower.layers.14.fc2", |
| "audio_tower.layers.15.self_attn.k_proj", |
| "audio_tower.layers.15.self_attn.v_proj", |
| "audio_tower.layers.15.self_attn.q_proj", |
| "audio_tower.layers.15.self_attn.out_proj", |
| "audio_tower.layers.15.fc1", |
| "audio_tower.layers.15.fc2", |
| "audio_tower.layers.16.self_attn.k_proj", |
| "audio_tower.layers.16.self_attn.v_proj", |
| "audio_tower.layers.16.self_attn.q_proj", |
| "audio_tower.layers.16.self_attn.out_proj", |
| "audio_tower.layers.16.fc1", |
| "audio_tower.layers.16.fc2", |
| "audio_tower.layers.17.self_attn.k_proj", |
| "audio_tower.layers.17.self_attn.v_proj", |
| "audio_tower.layers.17.self_attn.q_proj", |
| "audio_tower.layers.17.self_attn.out_proj", |
| "audio_tower.layers.17.fc1", |
| "audio_tower.layers.17.fc2", |
| "audio_tower.layers.18.self_attn.k_proj", |
| "audio_tower.layers.18.self_attn.v_proj", |
| "audio_tower.layers.18.self_attn.q_proj", |
| "audio_tower.layers.18.self_attn.out_proj", |
| "audio_tower.layers.18.fc1", |
| "audio_tower.layers.18.fc2", |
| "audio_tower.layers.19.self_attn.k_proj", |
| "audio_tower.layers.19.self_attn.v_proj", |
| "audio_tower.layers.19.self_attn.q_proj", |
| "audio_tower.layers.19.self_attn.out_proj", |
| "audio_tower.layers.19.fc1", |
| "audio_tower.layers.19.fc2", |
| "audio_tower.layers.20.self_attn.k_proj", |
| "audio_tower.layers.20.self_attn.v_proj", |
| "audio_tower.layers.20.self_attn.q_proj", |
| "audio_tower.layers.20.self_attn.out_proj", |
| "audio_tower.layers.20.fc1", |
| "audio_tower.layers.20.fc2", |
| "audio_tower.layers.21.self_attn.k_proj", |
| "audio_tower.layers.21.self_attn.v_proj", |
| "audio_tower.layers.21.self_attn.q_proj", |
| "audio_tower.layers.21.self_attn.out_proj", |
| "audio_tower.layers.21.fc1", |
| "audio_tower.layers.21.fc2", |
| "audio_tower.layers.22.self_attn.k_proj", |
| "audio_tower.layers.22.self_attn.v_proj", |
| "audio_tower.layers.22.self_attn.q_proj", |
| "audio_tower.layers.22.self_attn.out_proj", |
| "audio_tower.layers.22.fc1", |
| "audio_tower.layers.22.fc2", |
| "audio_tower.layers.23.self_attn.k_proj", |
| "audio_tower.layers.23.self_attn.v_proj", |
| "audio_tower.layers.23.self_attn.q_proj", |
| "audio_tower.layers.23.self_attn.out_proj", |
| "audio_tower.layers.23.fc1", |
| "audio_tower.layers.23.fc2", |
| "audio_tower.layers.24.self_attn.k_proj", |
| "audio_tower.layers.24.self_attn.v_proj", |
| "audio_tower.layers.24.self_attn.q_proj", |
| "audio_tower.layers.24.self_attn.out_proj", |
| "audio_tower.layers.24.fc1", |
| "audio_tower.layers.24.fc2", |
| "audio_tower.layers.25.self_attn.k_proj", |
| "audio_tower.layers.25.self_attn.v_proj", |
| "audio_tower.layers.25.self_attn.q_proj", |
| "audio_tower.layers.25.self_attn.out_proj", |
| "audio_tower.layers.25.fc1", |
| "audio_tower.layers.25.fc2", |
| "audio_tower.layers.26.self_attn.k_proj", |
| "audio_tower.layers.26.self_attn.v_proj", |
| "audio_tower.layers.26.self_attn.q_proj", |
| "audio_tower.layers.26.self_attn.out_proj", |
| "audio_tower.layers.26.fc1", |
| "audio_tower.layers.26.fc2", |
| "audio_tower.layers.27.self_attn.k_proj", |
| "audio_tower.layers.27.self_attn.v_proj", |
| "audio_tower.layers.27.self_attn.q_proj", |
| "audio_tower.layers.27.self_attn.out_proj", |
| "audio_tower.layers.27.fc1", |
| "audio_tower.layers.27.fc2", |
| "audio_tower.layers.28.self_attn.k_proj", |
| "audio_tower.layers.28.self_attn.v_proj", |
| "audio_tower.layers.28.self_attn.q_proj", |
| "audio_tower.layers.28.self_attn.out_proj", |
| "audio_tower.layers.28.fc1", |
| "audio_tower.layers.28.fc2", |
| "audio_tower.layers.29.self_attn.k_proj", |
| "audio_tower.layers.29.self_attn.v_proj", |
| "audio_tower.layers.29.self_attn.q_proj", |
| "audio_tower.layers.29.self_attn.out_proj", |
| "audio_tower.layers.29.fc1", |
| "audio_tower.layers.29.fc2", |
| "audio_tower.layers.30.self_attn.k_proj", |
| "audio_tower.layers.30.self_attn.v_proj", |
| "audio_tower.layers.30.self_attn.q_proj", |
| "audio_tower.layers.30.self_attn.out_proj", |
| "audio_tower.layers.30.fc1", |
| "audio_tower.layers.30.fc2", |
| "audio_tower.layers.31.self_attn.k_proj", |
| "audio_tower.layers.31.self_attn.v_proj", |
| "audio_tower.layers.31.self_attn.q_proj", |
| "audio_tower.layers.31.self_attn.out_proj", |
| "audio_tower.layers.31.fc1", |
| "audio_tower.layers.31.fc2", |
| "audio_tower.conv_out", |
| "audio_tower.proj1", |
| "audio_tower.proj2", |
| "visual.merger_list.0.mlp.0", |
| "visual.merger_list.0.mlp.2", |
| "visual.merger_list.1.mlp.0", |
| "visual.merger_list.1.mlp.2", |
| "visual.merger_list.2.mlp.0", |
| "visual.merger_list.2.mlp.2", |
| "visual.blocks.0.attn.qkv", |
| "visual.blocks.0.attn.proj", |
| "visual.blocks.0.mlp.linear_fc1", |
| "visual.blocks.0.mlp.linear_fc2", |
| "visual.blocks.1.attn.qkv", |
| "visual.blocks.1.attn.proj", |
| "visual.blocks.1.mlp.linear_fc1", |
| "visual.blocks.1.mlp.linear_fc2", |
| "visual.blocks.2.attn.qkv", |
| "visual.blocks.2.attn.proj", |
| "visual.blocks.2.mlp.linear_fc1", |
| "visual.blocks.2.mlp.linear_fc2", |
| "visual.blocks.3.attn.qkv", |
| "visual.blocks.3.attn.proj", |
| "visual.blocks.3.mlp.linear_fc1", |
| "visual.blocks.3.mlp.linear_fc2", |
| "visual.blocks.4.attn.qkv", |
| "visual.blocks.4.attn.proj", |
| "visual.blocks.4.mlp.linear_fc1", |
| "visual.blocks.4.mlp.linear_fc2", |
| "visual.blocks.5.attn.qkv", |
| "visual.blocks.5.attn.proj", |
| "visual.blocks.5.mlp.linear_fc1", |
| "visual.blocks.5.mlp.linear_fc2", |
| "visual.blocks.6.attn.qkv", |
| "visual.blocks.6.attn.proj", |
| "visual.blocks.6.mlp.linear_fc1", |
| "visual.blocks.6.mlp.linear_fc2", |
| "visual.blocks.7.attn.qkv", |
| "visual.blocks.7.attn.proj", |
| "visual.blocks.7.mlp.linear_fc1", |
| "visual.blocks.7.mlp.linear_fc2", |
| "visual.blocks.8.attn.qkv", |
| "visual.blocks.8.attn.proj", |
| "visual.blocks.8.mlp.linear_fc1", |
| "visual.blocks.8.mlp.linear_fc2", |
| "visual.blocks.9.attn.qkv", |
| "visual.blocks.9.attn.proj", |
| "visual.blocks.9.mlp.linear_fc1", |
| "visual.blocks.9.mlp.linear_fc2", |
| "visual.blocks.10.attn.qkv", |
| "visual.blocks.10.attn.proj", |
| "visual.blocks.10.mlp.linear_fc1", |
| "visual.blocks.10.mlp.linear_fc2", |
| "visual.blocks.11.attn.qkv", |
| "visual.blocks.11.attn.proj", |
| "visual.blocks.11.mlp.linear_fc1", |
| "visual.blocks.11.mlp.linear_fc2", |
| "visual.blocks.12.attn.qkv", |
| "visual.blocks.12.attn.proj", |
| "visual.blocks.12.mlp.linear_fc1", |
| "visual.blocks.12.mlp.linear_fc2", |
| "visual.blocks.13.attn.qkv", |
| "visual.blocks.13.attn.proj", |
| "visual.blocks.13.mlp.linear_fc1", |
| "visual.blocks.13.mlp.linear_fc2", |
| "visual.blocks.14.attn.qkv", |
| "visual.blocks.14.attn.proj", |
| "visual.blocks.14.mlp.linear_fc1", |
| "visual.blocks.14.mlp.linear_fc2", |
| "visual.blocks.15.attn.qkv", |
| "visual.blocks.15.attn.proj", |
| "visual.blocks.15.mlp.linear_fc1", |
| "visual.blocks.15.mlp.linear_fc2", |
| "visual.blocks.16.attn.qkv", |
| "visual.blocks.16.attn.proj", |
| "visual.blocks.16.mlp.linear_fc1", |
| "visual.blocks.16.mlp.linear_fc2", |
| "visual.blocks.17.attn.qkv", |
| "visual.blocks.17.attn.proj", |
| "visual.blocks.17.mlp.linear_fc1", |
| "visual.blocks.17.mlp.linear_fc2", |
| "visual.blocks.18.attn.qkv", |
| "visual.blocks.18.attn.proj", |
| "visual.blocks.18.mlp.linear_fc1", |
| "visual.blocks.18.mlp.linear_fc2", |
| "visual.blocks.19.attn.qkv", |
| "visual.blocks.19.attn.proj", |
| "visual.blocks.19.mlp.linear_fc1", |
| "visual.blocks.19.mlp.linear_fc2", |
| "visual.blocks.20.attn.qkv", |
| "visual.blocks.20.attn.proj", |
| "visual.blocks.20.mlp.linear_fc1", |
| "visual.blocks.20.mlp.linear_fc2", |
| "visual.blocks.21.attn.qkv", |
| "visual.blocks.21.attn.proj", |
| "visual.blocks.21.mlp.linear_fc1", |
| "visual.blocks.21.mlp.linear_fc2", |
| "visual.blocks.22.attn.qkv", |
| "visual.blocks.22.attn.proj", |
| "visual.blocks.22.mlp.linear_fc1", |
| "visual.blocks.22.mlp.linear_fc2", |
| "visual.blocks.23.attn.qkv", |
| "visual.blocks.23.attn.proj", |
| "visual.blocks.23.mlp.linear_fc1", |
| "visual.blocks.23.mlp.linear_fc2", |
| "visual.blocks.24.attn.qkv", |
| "visual.blocks.24.attn.proj", |
| "visual.blocks.24.mlp.linear_fc1", |
| "visual.blocks.24.mlp.linear_fc2", |
| "visual.blocks.25.attn.qkv", |
| "visual.blocks.25.attn.proj", |
| "visual.blocks.25.mlp.linear_fc1", |
| "visual.blocks.25.mlp.linear_fc2", |
| "visual.blocks.26.attn.qkv", |
| "visual.blocks.26.attn.proj", |
| "visual.blocks.26.mlp.linear_fc1", |
| "visual.blocks.26.mlp.linear_fc2", |
| "visual.merger.mlp.0", |
| "visual.merger.mlp.2", |
| "model.layers.0.mlp.gate", |
| "model.layers.1.mlp.gate", |
| "model.layers.2.mlp.gate", |
| "model.layers.3.mlp.gate", |
| "model.layers.4.mlp.gate", |
| "model.layers.5.mlp.gate", |
| "model.layers.6.mlp.gate", |
| "model.layers.7.mlp.gate", |
| "model.layers.8.mlp.gate", |
| "model.layers.9.mlp.gate", |
| "model.layers.10.mlp.gate", |
| "model.layers.11.mlp.gate", |
| "model.layers.12.mlp.gate", |
| "model.layers.13.mlp.gate", |
| "model.layers.14.mlp.gate", |
| "model.layers.15.mlp.gate", |
| "model.layers.16.mlp.gate", |
| "model.layers.17.mlp.gate", |
| "model.layers.18.mlp.gate", |
| "model.layers.19.mlp.gate", |
| "model.layers.20.mlp.gate", |
| "model.layers.21.mlp.gate", |
| "model.layers.22.mlp.gate", |
| "model.layers.23.mlp.gate", |
| "model.layers.24.mlp.gate", |
| "model.layers.25.mlp.gate", |
| "model.layers.26.mlp.gate", |
| "model.layers.27.mlp.gate", |
| "model.layers.28.mlp.gate", |
| "model.layers.29.mlp.gate", |
| "model.layers.30.mlp.gate", |
| "model.layers.31.mlp.gate", |
| "model.layers.32.mlp.gate", |
| "model.layers.33.mlp.gate", |
| "model.layers.34.mlp.gate", |
| "model.layers.35.mlp.gate", |
| "model.layers.36.mlp.gate", |
| "model.layers.37.mlp.gate", |
| "model.layers.38.mlp.gate", |
| "model.layers.39.mlp.gate", |
| "model.layers.40.mlp.gate", |
| "model.layers.41.mlp.gate", |
| "model.layers.42.mlp.gate", |
| "model.layers.43.mlp.gate", |
| "model.layers.44.mlp.gate", |
| "model.layers.45.mlp.gate", |
| "model.layers.46.mlp.gate", |
| "model.layers.47.mlp.gate", |
| "lm_head" |
| ], |
| "kv_cache_scheme": null, |
| "quant_method": "compressed-tensors", |
| "quantization_status": "compressed", |
| "sparsity_config": {}, |
| "transform_config": {}, |
| "version": "0.12.3.a20251203" |
| }, |
| "seconds_per_chunk": 2, |
| "text_config": { |
| "attention_bias": false, |
| "attention_dropout": 0.0, |
| "decoder_sparse_step": 1, |
| "head_dim": 128, |
| "hidden_act": "silu", |
| "hidden_size": 2048, |
| "initializer_range": 0.02, |
| "intermediate_size": 768, |
| "max_position_embeddings": 65536, |
| "mlp_only_layers": [], |
| "model_type": "qwen3_omni_moe_text", |
| "moe_intermediate_size": 768, |
| "norm_topk_prob": true, |
| "num_attention_heads": 32, |
| "num_experts": 128, |
| "num_experts_per_tok": 8, |
| "num_hidden_layers": 48, |
| "num_key_value_heads": 4, |
| "output_router_logits": false, |
| "rms_norm_eps": 1e-06, |
| "rope_scaling": { |
| "interleaved": true, |
| "mrope_interleaved": true, |
| "mrope_section": [ |
| 24, |
| 20, |
| 20 |
| ], |
| "rope_type": "default", |
| "type": "default" |
| }, |
| "rope_theta": 1000000, |
| "router_aux_loss_coef": 0.001, |
| "shared_expert_intermediate_size": 0, |
| "sliding_window": null, |
| "use_cache": true, |
| "use_qk_norm": true, |
| "use_sliding_window": false, |
| "vocab_size": 152064 |
| }, |
| "transformers_version": "4.57.0", |
| "user_token_id": 872, |
| "video_token_id": 151656, |
| "vision_config": { |
| "apply_vit_abs_pos_embed": true, |
| "deepstack_visual_indexes": [ |
| 8, |
| 16, |
| 24 |
| ], |
| "depth": 27, |
| "hidden_act": "gelu_pytorch_tanh", |
| "hidden_size": 1152, |
| "image_size": 768, |
| "in_channels": 3, |
| "in_chans": 3, |
| "initializer_range": 0.02, |
| "intermediate_size": 4304, |
| "model_type": "qwen3_omni_moe_vision_encoder", |
| "num_heads": 16, |
| "num_position_embeddings": 2304, |
| "out_hidden_size": 2048, |
| "patch_size": 16, |
| "spatial_merge_size": 2, |
| "spatial_patch_size": 16, |
| "temporal_patch_size": 2, |
| "tokens_per_second": 2 |
| }, |
| "vision_end_token_id": 151653, |
| "vision_start_token_id": 151652 |
| } |
