{
  "architecture": "vit_giant_patch16_lingbot",
  "num_classes": 0,
  "num_features": 1536,
  "global_pool": "avg",
  "pretrained_cfg": {
    "tag": "robbyant",
    "custom_load": false,
    "input_size": [
      3,
      512,
      512
    ],
    "fixed_input_size": true,
    "interpolation": "bilinear",
    "crop_pct": 1.0,
    "crop_mode": "squash",
    "mean": [
      0.485,
      0.456,
      0.406
    ],
    "std": [
      0.229,
      0.224,
      0.225
    ],
    "num_classes": 0,
    "pool_size": null,
    "first_conv": "patch_embed.proj",
    "classifier": "head",
    "license": "apache-2.0",
    "origin_url": "https://github.com/robbyant/lingbot-vision",
    "paper_ids": "arXiv:2607.05247"
  }
}