{
  "id": "auraface-v1",
  "identity": {
    "family": "auraface",
    "display_name": "AuraFace v1 (SCRFD-10GF + ArcFace-style ResNet100)",
    "hub_id": "fal/AuraFace-v1",
    "api_name": null,
    "revision": "af6d057c9b0ec4071d4c49c80e3539258798b609",
    "variants": [],
    "license": "Apache-2.0",
    "commercial_use": true,
    "commercial_use_note": "The model card states AuraFace was trained on commercially and publicly available data for commercial use, under Apache-2.0. The pack's SCRFD-10GF detector file comes from InsightFace's model zoo, so a detector trained for commercial use should replace it before a customer namespace relies on the pack.",
    "release_date": "2024-08",
    "paper_url": "https://arxiv.org/abs/1801.07698",
    "repo_url": "https://huggingface.co/fal/AuraFace-v1"
  },
  "task": {
    "category": "face",
    "modalities_in": [
      "image",
      "video"
    ],
    "modalities_out": [
      "boxes",
      "keypoints",
      "vector"
    ],
    "best_at": "512-d face identity vectors under a commercial license, loaded the same way as an InsightFace pack (FaceAnalysis with the pack's directory), so it drops in where buffalo_l runs.",
    "worst_at": "Separating a stranger from a match: on the face-check benchmark a person not in the index scored up to 0.418 against a true-match median of 0.55 (buffalo_l: 0.271 against 0.613), and the model card notes its training data may not cover every ethnicity well."
  },
  "io": {
    "input": {
      "image": {
        "det_size": "640x640",
        "align": "5-point affine to 112x112"
      },
      "video": "frames sampled by the extractor"
    },
    "output": {
      "boxes": "face bounding boxes with detection scores",
      "keypoints": "5 landmarks per face",
      "vector": {
        "dims": 512,
        "distance": "cosine",
        "normalized": true
      }
    }
  },
  "mixpeek_wiring": {
    "extractors": [],
    "feature_uris": [],
    "vector_indexes": [],
    "retriever_stages": [
      "feature_search"
    ],
    "composes_with": [
      "image or video -> SCRFD boxes -> aligned 112x112 crops -> AuraFace 512-d vector per face -> feature_search with a face photo"
    ]
  },
  "run": {
    "status": "plugin",
    "status_reason": "No Serve app or built-in extractor loads it yet. face_identity_extractor loads the buffalo_l pack by name; AuraFace needs the pack name (or a custom plugin) pointed at the AuraFace directory.",
    "serve_apps": [],
    "runtime": "insightface.app.FaceAnalysis(name=\"auraface\", allowed_modules=[\"detection\", \"recognition\"]) on onnxruntime, det_size=(640, 640)",
    "weights_format": "onnx",
    "trust_remote_code": false,
    "size_on_disk_mb": 265,
    "params_millions": null,
    "preprocessing": "SCRFD detection at 640x640, 5-point affine alignment to 112x112.",
    "postprocessing": "Recognition embedding per aligned face (512 dims), L2-normalized.",
    "serving_shape": "both",
    "batch_size": null,
    "serve_batch": null,
    "scale_to_zero": {
      "min_replicas": null,
      "max_replicas": null,
      "idle_timeout_s": null,
      "memory_gb": null,
      "num_cpus": null,
      "num_gpus": null,
      "cold_start_s": {
        "warm_node_s": null,
        "new_node_s": null
      },
      "cpu_fallback": true
    }
  },
  "cost": {
    "unmeasured": true,
    "measured": null,
    "note": "No committed throughput or cost measurement."
  },
  "evidence": {
    "unmeasured": false,
    "items": [
      {
        "dataset": "ip-check face-check benchmark, 82 photos of 4,978 indexed public figures (a different photo from the indexed portrait)",
        "metric": "recall@1, largest face per image, offline",
        "k": 1,
        "score": 0.927,
        "baseline": "insightface-buffalo-l on the same photos and index",
        "baseline_score": 0.976,
        "held_out": true,
        "date": "2026-11-16"
      }
    ]
  },
  "refs": [
    "hf://fal/AuraFace-v1@af6d057c9b0ec4071d4c49c80e3539258798b609"
  ],
  "formats": [
    {
      "ref": "hf://fal/AuraFace-v1@af6d057c9b0ec4071d4c49c80e3539258798b609",
      "format": "onnx",
      "file": "glintr100.onnx",
      "loader": "insightface FaceAnalysis model directory (glintr100.onnx recognition, scrfd_10g_bnkps.onnx detection)",
      "supported_today": true
    }
  ],
  "use_cases": [
    "face-search"
  ],
  "recommended": {
    "value": false,
    "for": [],
    "reason": null
  },
  "demos": [],
  "seo": {
    "title": "AuraFace v1: commercially licensed face recognition",
    "description": "AuraFace v1 is a face recognition model (Apache-2.0) for commercial face search. Can run on Mixpeek as a custom plugin.",
    "keywords": [
      "AuraFace",
      "auraface",
      "face recognition",
      "fal/AuraFace-v1",
      "face-search"
    ],
    "faqs": [
      {
        "q": "What is AuraFace v1 best at?",
        "a": "512-d face identity vectors under a commercial license, loaded the same way as an InsightFace pack, so it drops in where buffalo_l runs."
      },
      {
        "q": "What are the limits of AuraFace v1?",
        "a": "It separates strangers from matches less cleanly than buffalo_l (a stranger scored up to 0.418 against a true-match median of 0.55), and its training data may not cover every ethnicity well."
      },
      {
        "q": "Can AuraFace v1 be used commercially?",
        "a": "Yes. The recognition weights are Apache-2.0 and trained for commercial use; the bundled SCRFD detector comes from InsightFace's model zoo, so replace it before relying on the pack commercially."
      },
      {
        "q": "How does AuraFace v1 run on Mixpeek?",
        "a": "As a custom plugin today; no Serve app or built-in extractor loads it yet."
      },
      {
        "q": "What does AuraFace v1 output?",
        "a": "A 512-dimensional face vector per detected face, compared by cosine similarity."
      }
    ]
  },
  "last_verified": "2026-11-16",
  "sources": [
    "https://huggingface.co/fal/AuraFace-v1",
    "https://huggingface.co/api/models/fal/AuraFace-v1",
    "https://arxiv.org/abs/1801.07698"
  ],
  "templates": [],
  "usage": {
    "namespaces": null,
    "collections": null,
    "uncounted": true,
    "templates": 0,
    "counted_at": "2026-11-06T13:42:25Z"
  },
  "facets": {
    "status": "plugin",
    "category": "face",
    "family": "auraface",
    "license": "Apache-2.0",
    "commercial_use": true,
    "modalities_in": [
      "image",
      "video"
    ],
    "modalities_out": [
      "boxes",
      "keypoints",
      "vector"
    ],
    "use_cases": [
      "face-search"
    ],
    "recommended": false,
    "release_date": "2024-08",
    "params_millions": null,
    "size_on_disk_mb": 265,
    "vector_dims": 512,
    "vector_distance": "cosine",
    "weights_format": "onnx",
    "formats": [
      "onnx"
    ],
    "loadable_today": true,
    "trust_remote_code": false,
    "cpu_fallback": true,
    "num_gpus": null,
    "memory_gb": null,
    "min_replicas": null,
    "serving_shape": "both",
    "served": false,
    "ref_schemes": [
      "hf"
    ],
    "has_evidence": true,
    "cost_measured": false,
    "template_count": 0,
    "usage_namespaces": null,
    "usage_collections": null,
    "usage_uncounted": true,
    "has_demo": false,
    "rank_score": 2
  },
  "rank_score": 2
}
