{ "architectures": [ "HYV3VLForConditionalGeneration" ], "attention_bias": false, "attention_dropout": 0.0, "attention_head_dim": 128, "bos_token_id": 120000, "eod_token_id": 120020, "eos_token_id": 120025, "expert_hidden_dim": 768, "first_k_dense_replace": 1, "head_dim": 128, "hidden_act": "silu", "hidden_size": 2048, "image_start_token_id": 120118, "image_end_token_id": 120119, "image_token_id": 120120, "image_newline_token_id": 120121, "video_token_id": 120683, "video_start_token_id": 120122, "video_end_token_id": 120123, "initializer_range": 0.006, "intermediate_size": 6912, "max_position_embeddings": 32768, "mlp_bias": false, "model_type": "hy_v3_vl", "moe_router_enable_expert_bias": true, "moe_router_use_sigmoid": true, "n_group": null, "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, "num_shared_experts": 1, "org_vocab_size": 120818, "pad_id": 120002, "pad_token_id": 120002, "qk_norm": true, "rms_norm_eps": 1e-05, "rope_scaling": { "rope_type": "default", "mrope_interleaved": true, "mrope_section": [ 24, 20, 20 ] }, "rope_theta": 11158840, "route_norm": true, "router_scaling_factor": 2.826, "tie_word_embeddings": false, "transformers_version": "4.57.0", "use_cache": true, "use_grouped_mm": false, "vision_config": { "add_patchemb_bias": true, "attention_dropout": 0.0, "cat_extra_token": 0, "hidden_act": "gelu", "hidden_dropout": 0.0, "hidden_size": 1152, "img_max_token_num": 4096, "intermediate_size": 4304, "interpolate_mode": "bilinear", "max_image_size": 2048, "max_vit_seq_len": 16384, "num_attention_heads": 16, "num_channels": 3, "num_hidden_layers": 27, "out_hidden_size": 2048, "patch_size": 16, "rms_norm_eps": 1e-05, "temporal_patch_size": 1, "spatial_patch_size": 1, "spatial_merge_size": 2, "perceive_post_norm": false, "video_spatial_compression": false, "video_temporal_compression": false }, "vocab_size": 120818, "auto_map": { "AutoConfig": "configuration_hy_v3_vl.HYV3VLConfig", "AutoModel": "modeling_hy_v3_vl.HYV3VLModel", "AutoModelForCausalLM": "modeling_hy_v3_vl.HYV3VLForCausalLM", "AutoModelForImageTextToText": "modeling_hy_v3_vl.HYV3VLForConditionalGeneration", "AutoImageProcessor": "image_processing_hunyuan_vl.HunYuanVLImageProcessor", "AutoProcessor": "processing_hy_v3_vl.HYV3VLProcessor", "AutoVideoProcessor": "video_processing_hy_v3_vl.HYV3VLVideoProcessor" } }