config.json

{
  "add_classification_head": false,
  "architectures": [
    "UnifiedMoTForConditionalGeneration"
  ],
  "attention_bias": false,
  "attention_dropout": 0.0,
  "attention_head_dim": 128,
  "bos_token_id": 120000,
  "boundary_loss_weight": 0.0,
  "cla_share_factor": 2,
  "class_num": 0,
  "dense_list": [
    2048,
    0
  ],
  "dtype": "bfloat16",
  "eos_token_id": 120020,
  "flow_latent_placeholder_id": 120690,
  "flow_loss_weight": 1.0,
  "head_dim": 128,
  "hidden_act": "silu",
  "hidden_size": 2048,
  "im_end_id": 5,
  "im_newline_id": 11,
  "im_start_id": 4,
  "image_end_token_id": 120120,
  "image_start_token_id": 120119,
  "image_token_id": 120687,
  "initializer_range": 0.02,
  "intermediate_size": 6144,
  "latent_patch_size": 2,
  "latent_token_id": 120690,
  "mask_init_id": 12,
  "max_latent_size": 32,
  "max_position_embeddings": 262144,
  "mlp_bias": false,
  "model_type": "unified_mot",
  "norm_type": "rms",
  "num_attention_heads": 16,
  "num_hidden_layers": 32,
  "num_key_value_heads": 4,
  "org_vocab_size": 120818,
  "pad_id": 120002,
  "pad_token_id": 120002,
  "pool_type": "last",
  "pretraining_tp": 1,
  "rms_norm_eps": 1e-05,
  "rope_scaling": {
    "alpha": 1000.0,
    "beta_fast": 32,
    "beta_slow": 1,
    "factor": 1.0,
    "mscale": 1.0,
    "mscale_all_dim": 1.0,
    "type": "dynamic"
  },
  "rope_theta": 10000.0,
  "sep_token_id": 120007,
  "text_config": {
    "_name_or_path": "",
    "add_classification_head": false,
    "architectures": [
      "UnifiedMoTForConditionalGeneration"
    ],
    "attention_bias": false,
    "attention_dropout": 0.0,
    "attention_head_dim": 128,
    "cla_share_factor": 2,
    "class_num": 0,
    "dense_list": [
      2048,
      0
    ],
    "dtype": "bfloat16",
    "head_dim": 128,
    "hidden_act": "silu",
    "hidden_size": 2048,
    "im_end_id": 5,
    "im_newline_id": 11,
    "im_start_id": 4,
    "initializer_range": 0.02,
    "intermediate_size": 6144,
    "mask_init_id": 12,
    "max_position_embeddings": 262144,
    "mlp_bias": false,
    "mlp_g_init_noise_std": 0.0001,
    "mlp_g_intermediate_size": 12288,
    "model_type": "unified_mot",
    "norm_type": "rms",
    "num_attention_heads": 16,
    "num_hidden_layers": 32,
    "num_key_value_heads": 4,
    "org_vocab_size": 120818,
    "pad_id": 120002,
    "pool_type": "last",
    "pretraining_tp": 1,
    "rms_norm_eps": 1e-05,
    "rope_scaling": {
      "alpha": 1000.0,
      "beta_fast": 32,
      "beta_slow": 1,
      "factor": 1.0,
      "mscale": 1.0,
      "mscale_all_dim": 1.0,
      "type": "dynamic"
    },
    "rope_theta": 10000.0,
    "sep_token_id": 120007,
    "text_end_id": 7,
    "text_start_id": 6,
    "use_cache": true,
    "use_cla": false,
    "use_moe": false,
    "use_qk_norm": true,
    "use_rotary_pos_emb": true,
    "use_varlen_flash_attn": true,
    "vision_full_attention": false,
    "vocab_size": 120818
  },
  "text_end_id": 7,
  "text_loss_weight": 1.0,
  "text_start_id": 6,
  "timestep_shift": 1.0,
  "transformers_version": "4.57.0",
  "use_cache": true,
  "use_cla": false,
  "use_qk_norm": true,
  "use_rotary_pos_emb": true,
  "vae_image_downsample": 16,
  "vae_z_channels": 16,
  "video_end_token_id": 120123,
  "video_start_token_id": 120122,
  "vision_config": {
    "anyres_vit_max_image_size": 2048,
    "attention_dropout": 0.0,
    "cat_extra_token": 1,
    "dtype": "bfloat16",
    "hidden_act": "gelu",
    "hidden_size": 1152,
    "img_max_token_num": 4096,
    "intermediate_size": 4304,
    "interpolate_mode": "bilinear",
    "learnable_mlp_pooling_size": 0,
    "max_image_size": 2048,
    "max_vit_seq_len": 16384,
    "min_image_size": 512,
    "model_type": "hunyuan_vl_mot",
    "num_attention_heads": 16,
    "num_channels": 3,
    "num_hidden_layers": 27,
    "num_key_value_heads": 16,
    "out_hidden_size": 4096,
    "patch_size": 16,
    "perceive_post_norm": true,
    "perceive_pre_norm": true,
    "remove_prenorm": true,
    "resize_resolution": 2048,
    "rms_norm_eps": 1e-05,
    "spatial_merge_size": 2,
    "spatial_patch_size": 1,
    "temporal_patch_size": 1,
    "text_hidden_size": 2048,
    "video_max_image_size": 768,
    "video_min_image_size": 256,
    "vision_full_attention": false
  },
  "vocab_size": 120818
}