{
  "id": "clip-vit-b-16-datacomp-xl-s13b-b90k",
  "identity": {
    "family": "openclip",
    "display_name": "OpenCLIP ViT-B/16 DataComp-1B",
    "hub_id": "laion/CLIP-ViT-B-16-DataComp.XL-s13B-b90K",
    "api_name": null,
    "revision": null,
    "variants": [],
    "license": "MIT",
    "commercial_use": true,
    "commercial_use_note": "MIT weights; LAION notes the datasets are for research, the model license itself has no use restriction.",
    "release_date": "2023-05",
    "paper_url": "https://arxiv.org/abs/2304.14108",
    "repo_url": "https://github.com/mlfoundations/open_clip"
  },
  "task": {
    "category": "embedding",
    "modalities_in": [
      "image",
      "text"
    ],
    "modalities_out": [
      "vector"
    ],
    "best_at": "Strong B/16 trained on the filtered DataComp-1B set.",
    "worst_at": "English-centric text tower (77 tokens); weak on text in images and counting."
  },
  "io": {
    "input": {
      "image": {
        "resolution": "224x224",
        "color": "RGB",
        "normalize": "CLIP mean (0.4815, 0.4578, 0.4082), std (0.2686, 0.2613, 0.2758)"
      },
      "text": {
        "max_tokens": 77
      },
      "prompts": "'a photo of a {label}.'; open_clip tokenizer, 77 tokens"
    },
    "output": {
      "vector": {
        "dims": 512,
        "distance": "cosine",
        "type": "dense"
      },
      "note": "image and text vectors share one space; L2-normalize before cosine"
    }
  },
  "mixpeek_wiring": {
    "extractors": [],
    "feature_uris": [],
    "vector_indexes": [],
    "retriever_stages": [
      "feature_search",
      "mmr"
    ],
    "composes_with": [
      "video frames (VideoFrameExpander) -> image vectors -> feature_search with a text query",
      "detector boxes (RT-DETR, D-FINE) -> crops -> image vectors per object",
      "image vectors + E5 text vectors on captions -> feature_search fusion (dbsf)"
    ]
  },
  "run": {
    "status": "not_runnable",
    "status_reason": "Only open_clip_pytorch_model.bin (pickle) on the hub; a plugin needs a one-time safetensors conversion. Otherwise loads with open_clip.create_model_and_transforms('hf-hub:laion/CLIP-ViT-B-16-DataComp.XL-s13B-b90K').",
    "serve_apps": [],
    "runtime": "open_clip create_model_and_transforms('hf-hub:<id>')",
    "weights_format": "pickle",
    "trust_remote_code": false,
    "size_on_disk_mb": 599,
    "params_millions": 150,
    "preprocessing": "Resize and center-crop to 224, CLIP mean/std normalize; tokenize to 77.",
    "postprocessing": "L2-normalize; cosine similarity.",
    "serving_shape": "both",
    "batch_size": 32,
    "serve_batch": null,
    "scale_to_zero": {
      "min_replicas": 0,
      "max_replicas": 1,
      "idle_timeout_s": 600,
      "memory_gb": 2,
      "num_cpus": 1,
      "num_gpus": 0,
      "cold_start_s": {
        "warm_node_s": null,
        "new_node_s": null,
        "source": null
      },
      "cpu_fallback": true
    }
  },
  "cost": {
    "unmeasured": true,
    "measured": null,
    "note": "No committed measurement; plan row 3 measures it in the standalone image."
  },
  "evidence": {
    "unmeasured": true,
    "items": []
  },
  "refs": [
    "hf://laion/CLIP-ViT-B-16-DataComp.XL-s13B-b90K@d110532e8d4ff91c574ee60a342323f28468b287"
  ],
  "formats": [
    {
      "ref": "hf://laion/CLIP-ViT-B-16-DataComp.XL-s13B-b90K@d110532e8d4ff91c574ee60a342323f28468b287",
      "format": "pytorch_bin",
      "file": "open_clip_pytorch_model.bin",
      "loader": "torch.load (pickle); Mixpeek refuses pickle weights (shared/security/torch_pickle.py)",
      "supported_today": false
    }
  ],
  "use_cases": [
    "visual-search",
    "product-search"
  ],
  "recommended": {
    "value": false,
    "for": [],
    "reason": null
  },
  "demos": [],
  "seo": {
    "title": "OpenCLIP ViT-B/16 DataComp-1B: embedding model",
    "description": "OpenCLIP ViT-B/16 DataComp-1B is a image and text embedding model (MIT). Not runnable on Mixpeek today.",
    "keywords": [
      "OpenCLIP ViT-B/16 DataComp-1B",
      "openclip",
      "embedding",
      "laion/CLIP-ViT-B-16-DataComp.XL-s13B-b90K",
      "visual-search",
      "product-search"
    ],
    "faqs": [
      {
        "q": "What is OpenCLIP ViT-B/16 DataComp-1B best at?",
        "a": "Strong B/16 trained on the filtered DataComp-1B set."
      },
      {
        "q": "What are the limits of OpenCLIP ViT-B/16 DataComp-1B?",
        "a": "English-centric text tower (77 tokens); weak on text in images and counting."
      },
      {
        "q": "Can OpenCLIP ViT-B/16 DataComp-1B be used commercially?",
        "a": "Yes. MIT weights; LAION notes the datasets are for research, the model license itself has no use restriction."
      },
      {
        "q": "How does OpenCLIP ViT-B/16 DataComp-1B run on Mixpeek?",
        "a": "Only open_clip_pytorch_model.bin (pickle) on the hub; a plugin needs a one-time safetensors conversion. Otherwise loads with open_clip.create_model_and_transforms('hf-hub:laion/CLIP-ViT-B-16-DataComp.XL-s13B-b90K')."
      },
      {
        "q": "What does OpenCLIP ViT-B/16 DataComp-1B output?",
        "a": "A 512-dimensional dense vector, compared by cosine similarity."
      }
    ]
  },
  "last_verified": "2026-11-05",
  "sources": [
    "https://huggingface.co/laion/CLIP-ViT-B-16-DataComp.XL-s13B-b90K",
    "https://arxiv.org/abs/2304.14108",
    "https://github.com/mlfoundations/open_clip"
  ],
  "registry": null,
  "templates": [],
  "usage": {
    "namespaces": null,
    "collections": null,
    "uncounted": true,
    "templates": 0,
    "counted_at": null,
    "source": null
  },
  "facets": {
    "status": "not_runnable",
    "category": "embedding",
    "family": "openclip",
    "license": "MIT",
    "commercial_use": true,
    "modalities_in": [
      "image",
      "text"
    ],
    "modalities_out": [
      "vector"
    ],
    "use_cases": [
      "visual-search",
      "product-search"
    ],
    "recommended": false,
    "release_date": "2023-05",
    "params_millions": 150,
    "size_on_disk_mb": 599,
    "vector_dims": 512,
    "vector_distance": "cosine",
    "weights_format": "pickle",
    "formats": [
      "pytorch_bin"
    ],
    "loadable_today": false,
    "trust_remote_code": false,
    "cpu_fallback": true,
    "num_gpus": 0,
    "memory_gb": 2,
    "min_replicas": 0,
    "serving_shape": "both",
    "served": false,
    "ref_schemes": [
      "hf"
    ],
    "has_evidence": false,
    "cost_measured": false,
    "template_count": 0,
    "usage_namespaces": null,
    "usage_collections": null,
    "usage_uncounted": true,
    "has_demo": false,
    "rank_score": 0
  },
  "rank_score": 0
}