{
  "id": "internvideo2-5-chat-8b",
  "identity": {
    "family": "internvideo2.5",
    "display_name": "InternVideo2.5 Chat 8B",
    "hub_id": "OpenGVLab/InternVideo2_5_Chat_8B",
    "api_name": null,
    "revision": null,
    "variants": [
      "OpenGVLab/InternVideo2-Chat-8B (MIT, 2024)"
    ],
    "license": "Apache-2.0",
    "commercial_use": true,
    "commercial_use_note": "Apache-2.0 per the hub tag; the language model inside (InternLM/Qwen based InternVL) carries its own terms, check before shipping.",
    "release_date": "2025-01-22",
    "paper_url": "https://arxiv.org/abs/2501.12386",
    "repo_url": "https://github.com/OpenGVLab/InternVideo"
  },
  "task": {
    "category": "video_qa",
    "modalities_in": [
      "video",
      "text"
    ],
    "modalities_out": [
      "text"
    ],
    "best_at": "Long-video question answering and fine-grained temporal grounding with an open model (long and rich context modeling).",
    "worst_at": "8.4B parameters and remote code: GPU only, slow per clip; answers need checking against transcripts."
  },
  "io": {
    "input": {
      "video": {
        "frames": "up to hundreds of frames (dynamic sampling)",
        "resolution": "448x448 tiles"
      },
      "prompts": "free-form question",
      "max_tokens": "LLM context per the card"
    },
    "output": {
      "text": "free-form answer or JSON when prompted"
    }
  },
  "mixpeek_wiring": {
    "extractors": [],
    "feature_uris": [],
    "vector_indexes": [],
    "retriever_stages": [],
    "composes_with": [
      "scene splits -> InternVideo2.5 descriptions and tags -> E5 text vectors + BM25"
    ]
  },
  "run": {
    "status": "not_runnable",
    "status_reason": "Requires trust_remote_code (custom_code on the hub) and a 24 GB class GPU. Would run as a GPU batch plugin after vendoring the InternVL chat code.",
    "serve_apps": [],
    "runtime": "transformers AutoModel with trust_remote_code=True (InternVL chat)",
    "weights_format": "safetensors",
    "trust_remote_code": true,
    "size_on_disk_mb": 16802,
    "params_millions": 8401,
    "preprocessing": "Sample frames, tile to 448, normalize; build the chat prompt with frame tokens.",
    "postprocessing": "Decode text; parse JSON when requested.",
    "serving_shape": "batch",
    "batch_size": null,
    "serve_batch": null,
    "scale_to_zero": {
      "min_replicas": 0,
      "max_replicas": 1,
      "idle_timeout_s": 600,
      "memory_gb": 32,
      "num_cpus": 8,
      "num_gpus": 1,
      "cold_start_s": {
        "warm_node_s": null,
        "new_node_s": null,
        "source": null
      },
      "cpu_fallback": false
    }
  },
  "cost": {
    "unmeasured": true,
    "measured": null,
    "note": "No committed measurement; plan row 3 measures it in the standalone image."
  },
  "evidence": {
    "unmeasured": true,
    "items": []
  },
  "refs": [
    "hf://OpenGVLab/InternVideo2_5_Chat_8B@87680fe232af8681b55322f68541679b91f99df1"
  ],
  "formats": [
    {
      "ref": "hf://OpenGVLab/InternVideo2_5_Chat_8B@87680fe232af8681b55322f68541679b91f99df1",
      "format": "safetensors",
      "file": "4 files, e.g. model-00001-of-00004.safetensors",
      "loader": "transformers / open_clip from_pretrained; Mixpeek Hub import (api/models/huggingface.py)",
      "supported_today": true
    }
  ],
  "use_cases": [
    "video-understanding",
    "captioning"
  ],
  "recommended": {
    "value": false,
    "for": [],
    "reason": null
  },
  "demos": [],
  "seo": {
    "title": "InternVideo2.5 Chat 8B: video understanding model",
    "description": "InternVideo2.5 Chat 8B is a video and text video understanding model (Apache-2.0). Not runnable on Mixpeek today.",
    "keywords": [
      "InternVideo2.5 Chat 8B",
      "internvideo2.5",
      "video understanding",
      "OpenGVLab/InternVideo2_5_Chat_8B",
      "video-understanding",
      "captioning"
    ],
    "faqs": [
      {
        "q": "What is InternVideo2.5 Chat 8B best at?",
        "a": "Long-video question answering and fine-grained temporal grounding with an open model (long and rich context modeling)."
      },
      {
        "q": "What are the limits of InternVideo2.5 Chat 8B?",
        "a": "8.4B parameters and remote code: GPU only, slow per clip; answers need checking against transcripts."
      },
      {
        "q": "Can InternVideo2.5 Chat 8B be used commercially?",
        "a": "Yes. Apache-2.0 per the hub tag; the language model inside (InternLM/Qwen based InternVL) carries its own terms, check before shipping."
      },
      {
        "q": "How does InternVideo2.5 Chat 8B run on Mixpeek?",
        "a": "Requires trust_remote_code (custom_code on the hub) and a 24 GB class GPU. Would run as a GPU batch plugin after vendoring the InternVL chat code."
      }
    ]
  },
  "last_verified": "2026-11-05",
  "sources": [
    "https://huggingface.co/OpenGVLab/InternVideo2_5_Chat_8B",
    "https://arxiv.org/abs/2501.12386",
    "https://github.com/OpenGVLab/InternVideo"
  ],
  "registry": null,
  "templates": [],
  "usage": {
    "namespaces": null,
    "collections": null,
    "uncounted": true,
    "templates": 0,
    "counted_at": null,
    "source": null
  },
  "facets": {
    "status": "not_runnable",
    "category": "video_qa",
    "family": "internvideo2.5",
    "license": "Apache-2.0",
    "commercial_use": true,
    "modalities_in": [
      "video",
      "text"
    ],
    "modalities_out": [
      "text"
    ],
    "use_cases": [
      "video-understanding",
      "captioning"
    ],
    "recommended": false,
    "release_date": "2025-01-22",
    "params_millions": 8401,
    "size_on_disk_mb": 16802,
    "vector_dims": null,
    "vector_distance": null,
    "weights_format": "safetensors",
    "formats": [
      "safetensors"
    ],
    "loadable_today": true,
    "trust_remote_code": true,
    "cpu_fallback": false,
    "num_gpus": 1,
    "memory_gb": 32,
    "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
}