{
  "id": "yolo-world-v2",
  "identity": {
    "family": "yolo-world",
    "display_name": "YOLO-World v2 (Ultralytics)",
    "hub_id": "yolov8s-worldv2.pt",
    "api_name": null,
    "revision": null,
    "variants": [
      "yolov8m-worldv2.pt",
      "yolov8l-worldv2.pt",
      "yolov8x-worldv2.pt"
    ],
    "license": "AGPL-3.0",
    "commercial_use": false,
    "commercial_use_note": "AGPL-3.0: offering it over a network obliges publishing the whole application's source under AGPL; closed commercial use needs an Ultralytics Enterprise license. The original Tencent AILab repo is GPL-3.0.",
    "release_date": "2024-01-30",
    "paper_url": "https://arxiv.org/abs/2401.17270",
    "repo_url": "https://github.com/AILab-CVC/YOLO-World"
  },
  "task": {
    "category": "detection",
    "modalities_in": [
      "image",
      "text"
    ],
    "modalities_out": [
      "boxes",
      "labels",
      "scores"
    ],
    "best_at": "Open-vocabulary detection from a list of text class names, in real time; set classes per namespace without training.",
    "worst_at": "Rare or compositional phrases; accuracy below closed-set YOLO on COCO classes; license blocks closed commercial use."
  },
  "io": {
    "input": {
      "image": {
        "resolution": "640x640 letterboxed"
      },
      "prompts": "list of class names (CLIP text encoder), set once with set_classes"
    },
    "output": {
      "boxes": "xyxy pixels",
      "labels": "open vocabulary (the given class list)",
      "scores": "0..1"
    }
  },
  "mixpeek_wiring": {
    "extractors": [],
    "feature_uris": [],
    "vector_indexes": [],
    "retriever_stages": [],
    "composes_with": [
      "taxonomy labels -> YOLO-World class list -> boxes -> crops -> SigLIP",
      "detector boxes -> crops -> SigLIP vectors per object (one document per object, or a multi-vector per frame)"
    ]
  },
  "run": {
    "status": "not_runnable",
    "status_reason": "Blocked by AGPL-3.0 and pickle .pt weights. Would run as a custom extractor plugin that exports the checkpoint to ONNX once (ultralytics export format=onnx) and serves it with onnxruntime on CPU, emitting boxes, labels and crops. The class list is baked in at export.",
    "serve_apps": [],
    "runtime": "ultralytics YOLOWorld",
    "weights_format": "pickle",
    "trust_remote_code": false,
    "size_on_disk_mb": null,
    "params_millions": null,
    "preprocessing": "Letterbox to 640x640 (stride-aligned padding), RGB, scale to 0..1.",
    "postprocessing": "Confidence threshold then NMS.",
    "serving_shape": "both",
    "batch_size": null,
    "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": [
    "github://ultralytics/assets/yolov8s-worldv2.pt"
  ],
  "formats": [
    {
      "ref": "github://ultralytics/assets/yolov8s-worldv2.pt",
      "format": "pytorch_pt",
      "file": "yolov8s-worldv2.pt",
      "loader": "torch.load (pickle); Mixpeek refuses pickle weights (shared/security/torch_pickle.py)",
      "supported_today": false
    }
  ],
  "use_cases": [
    "object-detection",
    "brand-detection"
  ],
  "recommended": {
    "value": false,
    "for": [],
    "reason": null
  },
  "demos": [],
  "seo": {
    "title": "YOLO-World v2 (Ultralytics): object detection model",
    "description": "YOLO-World v2 (Ultralytics) is a image and text object detection model (AGPL-3.0). Not runnable on Mixpeek today.",
    "keywords": [
      "YOLO-World v2 (Ultralytics)",
      "yolo-world",
      "object detection",
      "yolov8s-worldv2.pt",
      "object-detection",
      "brand-detection"
    ],
    "faqs": [
      {
        "q": "What is YOLO-World v2 (Ultralytics) best at?",
        "a": "Open-vocabulary detection from a list of text class names, in real time; set classes per namespace without training."
      },
      {
        "q": "What are the limits of YOLO-World v2 (Ultralytics)?",
        "a": "Rare or compositional phrases; accuracy below closed-set YOLO on COCO classes; license blocks closed commercial use."
      },
      {
        "q": "Can YOLO-World v2 (Ultralytics) be used commercially?",
        "a": "No. AGPL-3.0: offering it over a network obliges publishing the whole application's source under AGPL; closed commercial use needs an Ultralytics Enterprise license. The original Tencent AILab repo is GPL-3.0."
      },
      {
        "q": "How does YOLO-World v2 (Ultralytics) run on Mixpeek?",
        "a": "Blocked by AGPL-3.0 and pickle .pt weights. Would run as a custom extractor plugin that exports the checkpoint to ONNX once (ultralytics export format=onnx) and serves it with onnxruntime on CPU, emitting boxes, labels and crops. The class list is baked in at export."
      }
    ]
  },
  "last_verified": "2026-11-05",
  "sources": [
    "https://docs.ultralytics.com/models/yolo-world/",
    "https://arxiv.org/abs/2401.17270",
    "https://github.com/AILab-CVC/YOLO-World"
  ],
  "registry": null,
  "templates": [],
  "usage": {
    "namespaces": null,
    "collections": null,
    "uncounted": true,
    "templates": 0,
    "counted_at": null,
    "source": null
  },
  "facets": {
    "status": "not_runnable",
    "category": "detection",
    "family": "yolo-world",
    "license": "AGPL-3.0",
    "commercial_use": false,
    "modalities_in": [
      "image",
      "text"
    ],
    "modalities_out": [
      "boxes",
      "labels",
      "scores"
    ],
    "use_cases": [
      "object-detection",
      "brand-detection"
    ],
    "recommended": false,
    "release_date": "2024-01-30",
    "params_millions": null,
    "size_on_disk_mb": null,
    "vector_dims": null,
    "vector_distance": null,
    "weights_format": "pickle",
    "formats": [
      "pytorch_pt"
    ],
    "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": [
      "github"
    ],
    "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
}