CompVis_LAION/config.json

{
  "library_name": "lucid",
  "architecture": "Stable Diffusion",
  "model_type": "stable_diffusion",
  "task": "base",
  "config": {
    "sample_size": 512,
    "in_channels": 3,
    "out_channels": 3,
    "act_fn": "silu",
    "num_train_timesteps": 1000,
    "beta_start": 0.00085,
    "beta_end": 0.012,
    "beta_schedule": "scaled_linear",
    "prediction_type": "epsilon",
    "latent_channels": 4,
    "downsample_factor": 8,
    "vae_block_out_channels": [
      128,
      256,
      512,
      512
    ],
    "vae_layers_per_block": 2,
    "unet_block_out_channels": [
      320,
      640,
      1280,
      1280
    ],
    "unet_layers_per_block": 2,
    "attention_head_dim": 8,
    "cross_attention_dim": 768,
    "context_length": 77,
    "norm_num_groups": 32
  },
  "preprocessing": {
    "resize": 512,
    "rescale": 0.00784313725490196,
    "mean": [
      0.5,
      0.5,
      0.5
    ],
    "std": [
      0.5,
      0.5,
      0.5
    ]
  },
  "weights": {
    "tag": "CompVis_LAION",
    "num_classes": 0,
    "license": "creativeml-openrail-m",
    "source": "https://huggingface.co/stable-diffusion-v1-5/stable-diffusion-v1-5",
    "sha256": "2fa36eeea24af004402d07d8c1ff6731f81119d69bdf952e5d0da5ad42f2860e",
    "file_size_mb": 3598.02
  },
  "id2label": {
    "0": "generative",
    "1": "diffusion",
    "2": "text-to-image"
  },
  "label2id": {
    "generative": 0,
    "diffusion": 1,
    "text-to-image": 2
  }
}