transformer/config.json

{
  "_class_name": "QwenImageTransformer2DModel",
  "_diffusers_version": "0.35.1",
  "attention_head_dim": 128,
  "axes_dims_rope": [
    16,
    56,
    56
  ],
  "guidance_embeds": false,
  "in_channels": 64,
  "joint_attention_dim": 3584,
  "num_attention_heads": 24,
  "num_layers": 60,
  "out_channels": 16,
  "patch_size": 2,
  "pooled_projection_dim": 768,
  "scale_rope": true,
  "tag_embedding_dim": 512,
  "tag_vocab_size": 18,
  "router_hidden_dim": 256,
  "moe_num_experts": 4,
  "runtime_moe_replace_from_layer": 50,
  "runtime_moe_replace_num_layers": 10,
  "runtime_moe_target": "transformer_blocks.*.img_mlp"
}