{
  "_class_name": "LLaDAImageSigVQModel",
  "_diffusers_version": "0.40.0.dev0",
  "attention_bias": true,
  "attention_dropout": 0.0,
  "codebook_embed_dim": 2048,
  "codebook_size": 16384,
  "hidden_size": 1536,
  "image_size": 2048,
  "in_channels": 3,
  "intermediate_size": 6144,
  "norm_eps": 1e-06,
  "num_attention_heads": 16,
  "num_hidden_layers": 40,
  "patch_size": 16,
  "semantic_embed_dim": 4096
}