{
  "id": "videoprism-lvt-base-f16r288",
  "identity": {
    "family": "videoprism",
    "display_name": "VideoPrism LvT Base (video-text, 16 frames, 288px)",
    "hub_id": "google/videoprism-lvt-base-f16r288",
    "api_name": null,
    "revision": null,
    "variants": [
      "google/videoprism-lvt-large-f8r288"
    ],
    "license": "Apache-2.0",
    "commercial_use": true,
    "commercial_use_note": "Apache-2.0 per the hub tag and repo.",
    "release_date": "2025-07",
    "paper_url": "https://arxiv.org/abs/2402.13217",
    "repo_url": "https://github.com/google-deepmind/videoprism"
  },
  "task": {
    "category": "embedding",
    "modalities_in": [
      "video",
      "text"
    ],
    "modalities_out": [
      "vector"
    ],
    "best_at": "Video-text retrieval and zero-shot classification with a paired text encoder.",
    "worst_at": "JAX/Flax only; text tower is English."
  },
  "io": {
    "input": {
      "video": {
        "frames": 16,
        "resolution": "288x288"
      },
      "text": {}
    },
    "output": {
      "vector": {
        "dims": "per model config (not verified here)",
        "distance": "cosine",
        "type": "dense"
      }
    }
  },
  "mixpeek_wiring": {
    "extractors": [],
    "feature_uris": [],
    "vector_indexes": [],
    "retriever_stages": [],
    "composes_with": [
      "scene splits (transcoding/multimodal extractor) -> clip-level vectors -> feature_search with a text query"
    ]
  },
  "run": {
    "status": "not_runnable",
    "status_reason": "Weights are Flax .npz arrays loaded by the videoprism JAX package; Mixpeek's runtime is PyTorch, so it needs JAX in the image or a port. Weights-only arrays, Apache-2.0, so no license block.",
    "serve_apps": [],
    "runtime": "videoprism (JAX/Flax)",
    "weights_format": "jax",
    "trust_remote_code": false,
    "size_on_disk_mb": 991,
    "params_millions": null,
    "preprocessing": "Sample 16 frames, resize to 288, scale to 0..1.",
    "postprocessing": "Global pooled embedding; L2-normalize for retrieval.",
    "serving_shape": "batch",
    "batch_size": null,
    "serve_batch": null,
    "scale_to_zero": {
      "min_replicas": 0,
      "max_replicas": 1,
      "idle_timeout_s": 600,
      "memory_gb": 6,
      "num_cpus": 2,
      "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://google/videoprism-lvt-base-f16r288@0119cca7d1fe61aa1d5718e1f71605703e1562e7"
  ],
  "formats": [
    {
      "ref": "hf://google/videoprism-lvt-base-f16r288@0119cca7d1fe61aa1d5718e1f71605703e1562e7",
      "format": "jax",
      "file": "flax_lvt_base_f16r288_repeated.npz",
      "loader": "JAX/Flax; not in Mixpeek's runtime",
      "supported_today": false
    }
  ],
  "use_cases": [
    "video-understanding",
    "action-recognition"
  ],
  "recommended": {
    "value": false,
    "for": [],
    "reason": null
  },
  "demos": [],
  "seo": {
    "title": "VideoPrism LvT Base (video-text, 16 frames, 288px): embedding model",
    "description": "VideoPrism LvT Base (video-text, 16 frames, 288px) is a video and text embedding model (Apache-2.0). Not runnable on Mixpeek today.",
    "keywords": [
      "VideoPrism LvT Base (video-text, 16 frames, 288px)",
      "videoprism",
      "embedding",
      "google/videoprism-lvt-base-f16r288",
      "video-understanding",
      "action-recognition"
    ],
    "faqs": [
      {
        "q": "What is VideoPrism LvT Base (video-text, 16 frames, 288px) best at?",
        "a": "Video-text retrieval and zero-shot classification with a paired text encoder."
      },
      {
        "q": "What are the limits of VideoPrism LvT Base (video-text, 16 frames, 288px)?",
        "a": "JAX/Flax only; text tower is English."
      },
      {
        "q": "Can VideoPrism LvT Base (video-text, 16 frames, 288px) be used commercially?",
        "a": "Yes. Apache-2.0 per the hub tag and repo."
      },
      {
        "q": "How does VideoPrism LvT Base (video-text, 16 frames, 288px) run on Mixpeek?",
        "a": "Weights are Flax .npz arrays loaded by the videoprism JAX package; Mixpeek's runtime is PyTorch, so it needs JAX in the image or a port. Weights-only arrays, Apache-2.0, so no license block."
      }
    ]
  },
  "last_verified": "2026-11-05",
  "sources": [
    "https://huggingface.co/google/videoprism-lvt-base-f16r288",
    "https://arxiv.org/abs/2402.13217",
    "https://github.com/google-deepmind/videoprism"
  ],
  "registry": null,
  "templates": [],
  "usage": {
    "namespaces": null,
    "collections": null,
    "uncounted": true,
    "templates": 0,
    "counted_at": null,
    "source": null
  },
  "facets": {
    "status": "not_runnable",
    "category": "embedding",
    "family": "videoprism",
    "license": "Apache-2.0",
    "commercial_use": true,
    "modalities_in": [
      "video",
      "text"
    ],
    "modalities_out": [
      "vector"
    ],
    "use_cases": [
      "video-understanding",
      "action-recognition"
    ],
    "recommended": false,
    "release_date": "2025-07",
    "params_millions": null,
    "size_on_disk_mb": 991,
    "vector_dims": null,
    "vector_distance": "cosine",
    "weights_format": "jax",
    "formats": [
      "jax"
    ],
    "loadable_today": false,
    "trust_remote_code": false,
    "cpu_fallback": true,
    "num_gpus": 0,
    "memory_gb": 6,
    "min_replicas": 0,
    "serving_shape": "batch",
    "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
}