{
  "schema_version": "2.0",
  "record_type": "dataset",
  "tier": "verified",
  "slug": "healthbench",
  "name": "HealthBench",
  "summary": {
    "en": "License: permissive (MIT); Mdexa hosting: link-only. first-hand data-scope statement(s) allow commercial use and redistribution",
    "zh": "许可证声明：宽松（MIT）；本站处理：只链接原出处。针对数据的声明允许商用和再分发。"
  },
  "systems": [
    "modern-medicine"
  ],
  "domains": [
    "clinical-medicine"
  ],
  "content_types": [
    "benchmark-exam"
  ],
  "content_description": {
    "en": "Contains the data for the HealthBench eval.",
    "zh": "包含 HealthBench 评估所需的数据。"
  },
  "languages": [
    "en"
  ],
  "modalities": [
    "text"
  ],
  "size": {
    "raw": null,
    "measures": []
  },
  "publisher": "openai",
  "year": {
    "value": null,
    "raw": null
  },
  "license": {
    "legacy_status": "pending",
    "spdx": "MIT",
    "commercial_use": "yes",
    "derivatives": "yes",
    "redistribution": "yes",
    "rationale": {
      "en": "first-hand data-scope statement(s) allow commercial use and redistribution",
      "zh": "针对数据的声明允许商用和再分发"
    },
    "evidence": [
      {
        "location": "hf-card-metadata",
        "location_detail": {
          "en": "Statement found in snapshot s09 (readme)",
          "zh": "在快照 s09（readme）中找到的声明"
        },
        "excerpt": "license: mit",
        "spdx": "MIT",
        "url": "https://huggingface.co/datasets/openai/healthbench",
        "verified_on": "2026-11-12",
        "snapshot_id": "s09",
        "observed": "page",
        "captured_by": "pipeline"
      }
    ],
    "license_conditions": [
      "notice-required"
    ]
  },
  "license_status": "permissive",
  "hosting": "link-only",
  "hosting_reason": {
    "en": "License permissive: first-hand data-scope statement(s) allow commercial use and redistribution. Link only.",
    "zh": "许可证宽松：针对数据的声明允许商用和再分发。只链接原出处。"
  },
  "signoff": null,
  "risk_flags": [],
  "provenance": {
    "source_description": null,
    "upstream": []
  },
  "known_issues": [],
  "preflight": {
    "status": "not-run"
  },
  "access": {
    "mdexa_download": null,
    "external_urls": [
      "https://huggingface.co/datasets/openai/healthbench"
    ],
    "notes": null
  },
  "citation": {
    "text": null,
    "references": []
  },
  "verification_level": "verified",
  "last_verified": "2026-11-14",
  "tags": [
    {
      "facet": "intended_use",
      "value": "evaluation",
      "basis": "inferred",
      "inferred_from": "card-content-description",
      "excerpt": "Contains the data for the [HealthBench eval](https://openai.com/index/healthbench/).",
      "snapshot_id": "s09"
    }
  ],
  "changelog": [
    {
      "date": "2026-11-14",
      "change": {
        "en": "Added via the automated pipeline and passed all checks (gates, verbatim quotes, rule engine, two independent extractions).",
        "zh": "经自动流水线收录并通过全部校验（门禁、逐字引用、判定引擎、两轮独立抽取）。"
      }
    }
  ],
  "access_recipe": {
    "verified_on": "2026-11-14",
    "platform": "huggingface",
    "repo_id": "openai/healthbench",
    "source": {
      "api_url": "https://huggingface.co/api/datasets/openai/healthbench",
      "fetched_on": "2026-11-14"
    },
    "access_state": "open",
    "files": [
      {
        "name": ".gitattributes",
        "url": "https://huggingface.co/datasets/openai/healthbench/resolve/40ee1968852fc57f625934251ac22be47077a8fb/.gitattributes",
        "size_bytes": 2747,
        "checksum": null,
        "format": "gitattributes"
      },
      {
        "name": "2025-05-07-06-14-12_oss_eval.jsonl",
        "url": "https://huggingface.co/datasets/openai/healthbench/resolve/40ee1968852fc57f625934251ac22be47077a8fb/2025-05-07-06-14-12_oss_eval.jsonl",
        "size_bytes": 60258154,
        "checksum": {
          "algorithm": "sha256",
          "value": "e99dd3c6372c10d6fcc5e385c5fae69d0dd40392dae56836ef9493ae324ecd2f"
        },
        "format": "jsonl"
      },
      {
        "name": "2025-05-07-06-14-12_oss_meta_eval.jsonl",
        "url": "https://huggingface.co/datasets/openai/healthbench/resolve/40ee1968852fc57f625934251ac22be47077a8fb/2025-05-07-06-14-12_oss_meta_eval.jsonl",
        "size_bytes": 136224466,
        "checksum": {
          "algorithm": "sha256",
          "value": "6a343ed3fcb6f28353c2c6c157ae30ddf2db874c947b27ba474ad54da9c5dbac"
        },
        "format": "jsonl"
      },
      {
        "name": "README.md",
        "url": "https://huggingface.co/datasets/openai/healthbench/resolve/40ee1968852fc57f625934251ac22be47077a8fb/README.md",
        "size_bytes": 232,
        "checksum": null,
        "format": "md"
      },
      {
        "name": "consensus_2025-05-09-20-00-46.jsonl",
        "url": "https://huggingface.co/datasets/openai/healthbench/resolve/40ee1968852fc57f625934251ac22be47077a8fb/consensus_2025-05-09-20-00-46.jsonl",
        "size_bytes": 36921504,
        "checksum": {
          "algorithm": "sha256",
          "value": "fabe37553327238928b05357d8b309737b5d426d6dee432a3376da618d0969d6"
        },
        "format": "jsonl"
      },
      {
        "name": "hard_2025-05-08-21-00-10.jsonl",
        "url": "https://huggingface.co/datasets/openai/healthbench/resolve/40ee1968852fc57f625934251ac22be47077a8fb/hard_2025-05-08-21-00-10.jsonl",
        "size_bytes": 12581564,
        "checksum": {
          "algorithm": "sha256",
          "value": "b0320430e5cd974e746585594c1dd10b5a3fc2aff9c72b26106c2c4a069d74e9"
        },
        "format": "jsonl"
      }
    ],
    "files_truncated": false,
    "commands": [
      {
        "command": "hf download openai/healthbench --repo-type dataset",
        "evidence": {
          "excerpt": "You can use the hf download command from the terminal to directly download files from the Hub.",
          "doc": "hf-download",
          "url": "https://huggingface.co/docs/huggingface_hub/guides/download",
          "fetched_on": "2026-11-14"
        }
      }
    ],
    "registration": null,
    "version": {
      "commit": "40ee1968852fc57f625934251ac22be47077a8fb",
      "last_modified": "2025-08-27T15:58:59.000Z"
    },
    "license_note": {
      "license_status": "permissive",
      "text": "Use according to the terms of the official source; this recipe does not say the data can be used freely."
    },
    "gaps": []
  },
  "_links": {
    "text": "Continue in this database. This record is one of 109 processed and verified records (datasets, data resources and open models). To find others, call `https://mcp.mdexa.com/v1/datasets?q=<words>` or connect the MCP server at `https://mcp.mdexa.com/mcp`; no key is needed for 20 calls per minute. Rows are under 1 KB; an empty result means \"not listed\". For a list of items, one batch check (free account) returns a report. Guide: /agents.md. Similar records: CBLUE (https://mdexa.com/datasets/cblue/); CMB (Chinese Medical Benchmark) (https://mdexa.com/datasets/cmb/); CMExam (https://mdexa.com/datasets/cmexam/).",
    "records": 109,
    "search": {
      "template": "https://mcp.mdexa.com/v1/datasets?q={words}",
      "example": "https://mcp.mdexa.com/v1/datasets?q=binding%20affinity"
    },
    "mcp": "https://mcp.mdexa.com/mcp",
    "anonymous_calls_per_minute": 20,
    "guide": "https://mdexa.com/agents.md",
    "faq": "https://mdexa.com/developers/faq/",
    "similar": [
      {
        "name": "CBLUE",
        "url": "https://mdexa.com/datasets/cblue/",
        "json": "https://mdexa.com/datasets/cblue.json"
      },
      {
        "name": "CMB (Chinese Medical Benchmark)",
        "url": "https://mdexa.com/datasets/cmb/",
        "json": "https://mdexa.com/datasets/cmb.json"
      },
      {
        "name": "CMExam",
        "url": "https://mdexa.com/datasets/cmexam/",
        "json": "https://mdexa.com/datasets/cmexam.json"
      }
    ]
  }
}