{
  "_class_name": "HunyuanImageTransformer2DModel",
  "_diffusers_version": "0.36.0.dev0",
  "attention_head_dim": 128,
  "guidance_embeds": false,
  "in_channels": 64,
  "mlp_ratio": 4.0,
  "num_attention_heads": 28,
  "num_layers": 20,
  "num_refiner_layers": 2,
  "num_single_layers": 40,
  "out_channels": 64,
  "patch_size": [
    1,
    1
  ],
  "qk_norm": "rms_norm",
  "rope_axes_dim": [
    64,
    64
  ],
  "rope_theta": 256.0,
  "text_embed_2_dim": 1472,
  "text_embed_dim": 3584
}