{
  "scope": "genuine released official CLIP ViT-B/32; two photos, no benchmark",
  "provenance": {
    "model": "CLIP ViT-B/32",
    "sourceRevision": "d05afc436d78f1c48dc0dbf8e5980a9d471f35f6",
    "sourceHashes": {
      "LICENSE": {
        "url": "https://raw.githubusercontent.com/openai/CLIP/d05afc436d78f1c48dc0dbf8e5980a9d471f35f6/LICENSE",
        "sha256": "987e63b32f6c89ff5160e429458a872ff048e6860b590a3912e938f9da8f14db",
        "bytes": 1064
      },
      "README.md": {
        "url": "https://raw.githubusercontent.com/openai/CLIP/d05afc436d78f1c48dc0dbf8e5980a9d471f35f6/README.md",
        "sha256": "f82c5c75e140532eb37a7d943921e1bdd57d740c6b40d0030985bc5b5d11d6f1",
        "bytes": 7306
      },
      "clip/model.py": {
        "url": "https://raw.githubusercontent.com/openai/CLIP/d05afc436d78f1c48dc0dbf8e5980a9d471f35f6/clip/model.py",
        "sha256": "9902cbe5ee90a1da2aa3e6f043e8a23dc1f8831193b963785c9af03d5c7bef2c",
        "bytes": 17374
      },
      "clip/simple_tokenizer.py": {
        "url": "https://raw.githubusercontent.com/openai/CLIP/d05afc436d78f1c48dc0dbf8e5980a9d471f35f6/clip/simple_tokenizer.py",
        "sha256": "d1b09f10ed3d1e2c343619cb98cbfacb92363dd8607526a43aa7fe2d60d15bad",
        "bytes": 4632
      },
      "clip/clip.py": {
        "url": "https://raw.githubusercontent.com/openai/CLIP/d05afc436d78f1c48dc0dbf8e5980a9d471f35f6/clip/clip.py",
        "sha256": "9540f200fbf8145479fa655382a56dab048d238cc698b9cbd8df3b6d86d3f1b6",
        "bytes": 9644
      },
      "clip/bpe_simple_vocab_16e6.txt.gz": {
        "url": "https://raw.githubusercontent.com/openai/CLIP/d05afc436d78f1c48dc0dbf8e5980a9d471f35f6/clip/bpe_simple_vocab_16e6.txt.gz",
        "sha256": "924691ac288e54409236115652ad4aa250f48203de50a9e4722a6ecd48d6804a",
        "bytes": 1356917
      }
    },
    "checkpointUrl": "https://openaipublic.azureedge.net/clip/models/40d365715913c9da98579312b702a82c18be219cc2a73407c4526f58eba950af/ViT-B-32.pt",
    "checkpointSha256": "40d365715913c9da98579312b702a82c18be219cc2a73407c4526f58eba950af",
    "programSha256": "1bed536cc9e35bcd3b24edbde3cf4b8a7af4fc1b41cffa66c417d0178af210f0",
    "exportSha256": "3497a809e68ed6067c22f368f990f20008476f59c8f8b004be6d05b2ecf4c7d4",
    "sourceArchiveSha256": "25f08313679a9a6766e60040e245f0742e5e0b1027b858f7bc8183b22bdbfb37",
    "python": "3.13.13",
    "torch": "2.12.0+cpu",
    "torchvision": "0.27.0+cpu",
    "numpy": "2.1.2",
    "Pillow": "12.2.0",
    "device": "cpu",
    "dtype": "float32",
    "threads": 2,
    "parameterCount": 151277313,
    "reproductionCommand": "python clip_export.py --source-dir clip_native --checkpoint ViT-B-32.pt --output clip-demo",
    "adapters": "Native model unchanged; native byte map and BPE method with printable ASCII token splitting/cleanup; no ftfy or regex normalization."
  },
  "preprocessing": {
    "shortSide": 224,
    "centerCrop": [
      224,
      224
    ],
    "interpolation": "PIL bicubic",
    "rgbAfterCrop": true,
    "mean": [
      0.48145466,
      0.4578275,
      0.40821073
    ],
    "std": [
      0.26862954,
      0.26130258,
      0.27577711
    ]
  },
  "classes": [
    "cat",
    "dog",
    "cup of coffee",
    "bowl of soup",
    "car"
  ],
  "templates": [
    "a photo of a {}.",
    "a close-up photo of a {}."
  ],
  "prompts": [
    "a photo of a cat.",
    "a close-up photo of a cat.",
    "a photo of a dog.",
    "a close-up photo of a dog.",
    "a photo of a cup of coffee.",
    "a close-up photo of a cup of coffee.",
    "a photo of a bowl of soup.",
    "a close-up photo of a bowl of soup.",
    "a photo of a car.",
    "a close-up photo of a car."
  ],
  "tokens": [
    [
      49406,
      320,
      1125,
      539,
      320,
      2368,
      269,
      49407,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0
    ],
    [
      49406,
      320,
      2660,
      268,
      705,
      1125,
      539,
      320,
      2368,
      269,
      49407,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0
    ],
    [
      49406,
      320,
      1125,
      539,
      320,
      1929,
      269,
      49407,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0
    ],
    [
      49406,
      320,
      2660,
      268,
      705,
      1125,
      539,
      320,
      1929,
      269,
      49407,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0
    ],
    [
      49406,
      320,
      1125,
      539,
      320,
      1937,
      539,
      2453,
      269,
      49407,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0
    ],
    [
      49406,
      320,
      2660,
      268,
      705,
      1125,
      539,
      320,
      1937,
      539,
      2453,
      269,
      49407,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0
    ],
    [
      49406,
      320,
      1125,
      539,
      320,
      3814,
      539,
      7077,
      269,
      49407,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0
    ],
    [
      49406,
      320,
      2660,
      268,
      705,
      1125,
      539,
      320,
      3814,
      539,
      7077,
      269,
      49407,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0
    ],
    [
      49406,
      320,
      1125,
      539,
      320,
      1615,
      269,
      49407,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0
    ],
    [
      49406,
      320,
      2660,
      268,
      705,
      1125,
      539,
      320,
      1615,
      269,
      49407,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0,
      0
    ]
  ],
  "contextLength": 77,
  "vocabSize": 49408,
  "logitScaleLog": 4.605170249938965,
  "logitScaleExp": 100.00000762939453,
  "scalePolicy": "raw released evaluation exp(logit_scale), no training cap",
  "conditionality": "Softmax is conditional on these candidate prototypes and templates; uncalibrated.",
  "photos": [
    {
      "id": "cat",
      "title": "Chelsea the cat",
      "originalSize": [
        300,
        451
      ],
      "originalImage": "source-cat.png",
      "inputImage": "input-cat.png",
      "inputImageSha256": "876634f244f850e283e2f8dc581988d5a0ba75c289d757ae10c3851c8b7f23cc",
      "arrays": "arrays-cat.npz",
      "arraysSha256": "973e218ff84235d5159ca8648ac0074fc42dc78fd8b70324c5705e4c56c06f2c",
      "inputTensorSha256": "59ba65f37c29915984addd8e4d8856f9c1579c284af0ef0260bc4f25eadbbab9",
      "source": {
        "url": "https://raw.githubusercontent.com/scikit-image/scikit-image/v0.25.2/skimage/data/chelsea.png",
        "sha256": "596aa1e7cb875eb79f437e310381d26b338a81c2da23439704a73c4651e8c4bb",
        "author": "Stefan van der Walt",
        "license": "CC0",
        "licenseUrl": "https://github.com/scikit-image/scikit-image/blob/v0.25.2/skimage/data/_fetchers.py"
      },
      "cosine": [
        0.30230581760406494,
        0.24639800190925598,
        0.22371649742126465,
        0.22711583971977234,
        0.21089980006217957
      ],
      "logits": [
        30.23058319091797,
        24.639802932739258,
        22.37165069580078,
        22.711585998535156,
        21.089981079101562
      ],
      "softmax": [
        0.9952545166015625,
        0.0037144040688872337,
        0.00038445289828814566,
        0.0005401012022048235,
        0.00010671392374206334
      ],
      "nativePromptParityMaxError": 0.0,
      "nativeCandidateArithmeticMaxError": 7.62939453125e-06
    },
    {
      "id": "coffee",
      "title": "Cup of coffee",
      "originalSize": [
        400,
        600
      ],
      "originalImage": "source-coffee.png",
      "inputImage": "input-coffee.png",
      "inputImageSha256": "0f747fad739c4c2f206613d53c81ef3a66ea20aab884be8a39c134195286ff1f",
      "arrays": "arrays-coffee.npz",
      "arraysSha256": "0a9a9b6efff499426e961bdc197709095b2fca486a0011a5a3b664a374ca0754",
      "inputTensorSha256": "8a233b27be63ef4b34178acb019b9c766c33972159ffc4720efbbd0da622de75",
      "source": {
        "url": "https://raw.githubusercontent.com/scikit-image/scikit-image/v0.25.2/skimage/data/coffee.png",
        "sha256": "cc02f8ca188b167c775a7101b5d767d1e71792cf762c33d6fa15a4599b5a8de7",
        "author": "Rachel Michetti",
        "license": "CC0",
        "licenseUrl": "https://github.com/scikit-image/scikit-image/blob/v0.25.2/skimage/data/_fetchers.py"
      },
      "cosine": [
        0.19442296028137207,
        0.18607407808303833,
        0.3048168122768402,
        0.24276506900787354,
        0.195073664188385
      ],
      "logits": [
        19.442296981811523,
        18.60740852355957,
        30.4816837310791,
        24.276508331298828,
        19.507368087768555
      ],
      "softmax": [
        1.602366683073342e-05,
        6.953036063350737e-06,
        0.9979451298713684,
        0.002014806028455496,
        1.710101787466556e-05
      ],
      "nativePromptParityMaxError": 0.0,
      "nativeCandidateArithmeticMaxError": 7.62939453125e-06
    }
  ]
}