{
  "id": "yolox-s",
  "identity": {
    "family": "yolox",
    "display_name": "YOLOX-S",
    "hub_id": "yolox_s (Megvii release asset)",
    "api_name": null,
    "revision": null,
    "variants": [
      "yolox_nano",
      "yolox_tiny",
      "yolox_m",
      "yolox_l",
      "yolox_x"
    ],
    "license": "Apache-2.0",
    "commercial_use": true,
    "commercial_use_note": "Apache-2.0 weights and code; no use restrictions.",
    "release_date": "2021-07",
    "paper_url": "https://arxiv.org/abs/2107.08430",
    "repo_url": "https://github.com/Megvii-BaseDetection/YOLOX"
  },
  "task": {
    "category": "detection",
    "modalities_in": [
      "image"
    ],
    "modalities_out": [
      "boxes",
      "labels",
      "scores"
    ],
    "best_at": "Mature Apache-2.0 YOLO-style detector with official ONNX exports; easy CPU deployment.",
    "worst_at": "Older (2021): lower accuracy than D-FINE or RT-DETRv2 at equal cost; closed COCO label set."
  },
  "io": {
    "input": {
      "image": {
        "resolution": "640x640 letterboxed",
        "color": "BGR, no normalization (reference demo)"
      },
      "prompts": "none"
    },
    "output": {
      "boxes": "xyxy pixels after grid decode",
      "labels": "COCO 80 classes",
      "scores": "objectness x class"
    }
  },
  "mixpeek_wiring": {
    "extractors": [],
    "feature_uris": [],
    "vector_indexes": [],
    "retriever_stages": [],
    "composes_with": [
      "detector boxes -> crops -> SigLIP vectors per object (one document per object, or a multi-vector per frame)",
      "labels -> payload field -> attribute filter in a retriever",
      "scene splits -> keyframes -> detector -> object documents with frame timestamps"
    ]
  },
  "run": {
    "status": "plugin",
    "status_reason": "Not served. Would run as a custom extractor plugin with onnxruntime on CPU using the official yolox_s.onnx release asset (no Python model code needed).",
    "serve_apps": [],
    "runtime": "onnxruntime (official ONNX release)",
    "weights_format": "onnx",
    "trust_remote_code": false,
    "size_on_disk_mb": null,
    "params_millions": 9,
    "preprocessing": "Letterbox to 640 with padding value 114.",
    "postprocessing": "Decode grid outputs, score threshold, 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": 1,
      "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://Megvii-BaseDetection/YOLOX/yolox_s.onnx@0.1.1rc0"
  ],
  "formats": [
    {
      "ref": "github://Megvii-BaseDetection/YOLOX/yolox_s.onnx@0.1.1rc0",
      "format": "onnx",
      "file": "yolox_s.onnx",
      "loader": "onnxruntime; Mixpeek model upload (api/models/loadability.py, ONNX IR <= 11)",
      "supported_today": true
    }
  ],
  "use_cases": [
    "object-detection"
  ],
  "recommended": {
    "value": false,
    "for": [],
    "reason": null
  },
  "demos": [],
  "seo": {
    "title": "YOLOX-S: object detection model",
    "description": "YOLOX-S is a image object detection model (Apache-2.0). Can run on Mixpeek as a custom plugin.",
    "keywords": [
      "YOLOX-S",
      "yolox",
      "object detection",
      "yolox_s (Megvii release asset)",
      "object-detection"
    ],
    "faqs": [
      {
        "q": "What is YOLOX-S best at?",
        "a": "Mature Apache-2.0 YOLO-style detector with official ONNX exports; easy CPU deployment."
      },
      {
        "q": "What are the limits of YOLOX-S?",
        "a": "Older (2021): lower accuracy than D-FINE or RT-DETRv2 at equal cost; closed COCO label set."
      },
      {
        "q": "Can YOLOX-S be used commercially?",
        "a": "Yes. Apache-2.0 weights and code; no use restrictions."
      },
      {
        "q": "How does YOLOX-S run on Mixpeek?",
        "a": "Not served. Would run as a custom extractor plugin with onnxruntime on CPU using the official yolox_s.onnx release asset (no Python model code needed)."
      }
    ]
  },
  "last_verified": "2026-11-05",
  "sources": [
    "https://github.com/Megvii-BaseDetection/YOLOX",
    "https://arxiv.org/abs/2107.08430",
    "https://github.com/Megvii-BaseDetection/YOLOX/tree/main/demo/ONNXRuntime"
  ],
  "registry": null,
  "templates": [],
  "usage": {
    "namespaces": null,
    "collections": null,
    "uncounted": true,
    "templates": 0,
    "counted_at": null,
    "source": null
  },
  "facets": {
    "status": "plugin",
    "category": "detection",
    "family": "yolox",
    "license": "Apache-2.0",
    "commercial_use": true,
    "modalities_in": [
      "image"
    ],
    "modalities_out": [
      "boxes",
      "labels",
      "scores"
    ],
    "use_cases": [
      "object-detection"
    ],
    "recommended": false,
    "release_date": "2021-07",
    "params_millions": 9,
    "size_on_disk_mb": null,
    "vector_dims": null,
    "vector_distance": null,
    "weights_format": "onnx",
    "formats": [
      "onnx"
    ],
    "loadable_today": true,
    "trust_remote_code": false,
    "cpu_fallback": true,
    "num_gpus": 0,
    "memory_gb": 1,
    "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": 1
  },
  "rank_score": 1
}