{
  "_class_name": "ControlNetXSModel",
  "_diffusers_version": "0.24.0.dev0",
  "base_model_channel_sizes": {
    "down": [
      [
        4,
        320
      ],
      [
        320,
        320
      ],
      [
        320,
        320
      ],
      [
        320,
        320
      ],
      [
        320,
        640
      ],
      [
        640,
        640
      ],
      [
        640,
        640
      ],
      [
        640,
        1280
      ],
      [
        1280,
        1280
      ]
    ],
    "mid": [
      [
        1280,
        1280
      ]
    ],
    "up": [
      [
        2560,
        1280
      ],
      [
        2560,
        1280
      ],
      [
        1920,
        1280
      ],
      [
        1920,
        640
      ],
      [
        1280,
        640
      ],
      [
        960,
        640
      ],
      [
        960,
        320
      ],
      [
        640,
        320
      ],
      [
        640,
        320
      ]
    ]
  },
  "block_out_channels": [
    32,
    64,
    128
  ],
  "conditioning_channels": 3,
  "conditioning_embedding_out_channels": [
    16,
    32,
    96,
    256
  ],
  "controlnet_conditioning_channel_order": "rgb",
  "cross_attention_dim": 2048,
  "down_block_types": [
    "DownBlock2D",
    "CrossAttnDownBlock2D",
    "CrossAttnDownBlock2D"
  ],
  "learn_embedding": true,
  "norm_num_groups": 32,
  "num_attention_heads": [
    1,
    1,
    2
  ],
  "sample_size": 128,
  "time_embedding_dim": 1280,
  "time_embedding_input_dim": 320,
  "time_embedding_mix": 0.95,
  "transformer_layers_per_block": [
    1,
    2,
    10
  ],
  "up_block_types": [
    "CrossAttnUpBlock2D",
    "CrossAttnUpBlock2D",
    "UpBlock2D"
  ],
  "upcast_attention": null
}