{
  "schema_version": "2.0",
  "record_type": "dataset",
  "tier": "verified",
  "slug": "huatuo-26m",
  "name": "Huatuo-26M",
  "summary": {
    "zh": "许可证声明：宽松（Apache-2.0）；本站处理：只链接原出处。许可证宽松，但有来源风险：内容抓取自网站，原站条款未说明；源自真实患者文本；只链接原出处。",
    "en": "License: permissive (Apache-2.0); Mdexa hosting: link-only. License is permissive, but source risk: content scraped from websites whose terms are not stated; derived from real patient text; link only."
  },
  "systems": [
    "modern-medicine"
  ],
  "domains": [
    "clinical-medicine"
  ],
  "content_types": [
    "qa-dialogue"
  ],
  "content_description": {
    "en": "QA corpus (encyclopedia, knowledge graph, consultations)",
    "zh": "问答语料（百科、知识图谱、问诊）"
  },
  "languages": [
    "zh"
  ],
  "modalities": [
    "text"
  ],
  "size": {
    "raw": "超过2,600万条医疗问答对",
    "measures": [
      {
        "value": 26000000,
        "unit": "qa-pairs",
        "qualifier": "at-least",
        "scope": "total"
      }
    ]
  },
  "publisher": "FreedomIntelligence / 深圳市大数据研究院（arXiv 2305.01526）",
  "year": {
    "value": 2023,
    "raw": "2023"
  },
  "expert_review": {
    "en": "Not stated",
    "zh": "未说明"
  },
  "license": {
    "legacy_status": "distributable",
    "spdx": "Apache-2.0",
    "commercial_use": "yes",
    "derivatives": "yes",
    "redistribution": "yes",
    "rationale": {
      "en": "README states Apache 2.0",
      "zh": "README 写明 Apache 2.0"
    },
    "evidence": [
      {
        "location": "readme",
        "location_detail": {
          "en": "README license statement",
          "zh": "README 许可声明"
        },
        "excerpt": "The Huatuo-26M dataset is licensed under Apache 2.0.",
        "spdx": "Apache-2.0",
        "url": "https://github.com/FreedomIntelligence/Huatuo-26M",
        "verified_on": "2026-11-12",
        "snapshot_id": "s01",
        "observed": "page",
        "secondhand": false
      }
    ]
  },
  "license_status": "permissive",
  "hosting": "link-only",
  "hosting_reason": {
    "zh": "许可证宽松，但有来源风险：内容抓取自网站，原站条款未说明；源自真实患者文本；只链接原出处。",
    "en": "License is permissive, but source risk: content scraped from websites whose terms are not stated; derived from real patient text; link only."
  },
  "signoff": null,
  "risk_flags": [
    "scraped",
    "real-patient-text"
  ],
  "provenance": {
    "source_description": {
      "en": "Online medical encyclopedias, online medical knowledge bases and online consultation records (consultation answers given as URLs)",
      "zh": "在线医学百科、在线医学知识库、在线问诊记录（问诊答案以URL形式给出）"
    },
    "upstream": []
  },
  "known_issues": [
    {
      "en": "Content scraped from websites; the original sites' copyright and redistribution terms are not stated",
      "zh": "内容抓取自网站，原站点对内容的版权与再分发条款未说明"
    },
    {
      "en": "Consultation answers are provided as URLs",
      "zh": "问诊答案以URL形式提供"
    }
  ],
  "preflight": {
    "status": "not-run"
  },
  "access": {
    "mdexa_download": null,
    "external_urls": [
      "https://github.com/FreedomIntelligence/Huatuo-26M"
    ],
    "notes": null
  },
  "citation": {
    "text": null,
    "bibtex": null,
    "references": [
      "arXiv:2305.01526"
    ]
  },
  "verification_level": "verified",
  "last_verified": "2026-11-12",
  "tags": [
    {
      "facet": "intended_use",
      "value": "pretraining",
      "basis": "author-stated",
      "excerpt": "- As pre-training data for language model (LM):",
      "snapshot_id": "s01"
    },
    {
      "facet": "intended_use",
      "value": "fine-tuning",
      "basis": "author-stated",
      "excerpt": "- As fine-tuning data for Medical LLM:",
      "snapshot_id": "s01"
    },
    {
      "facet": "intended_use",
      "value": "evaluation",
      "basis": "author-stated",
      "excerpt": "The following is the huatuo test set we used in the paper, which consists of random sampling of data from multiple sources.",
      "snapshot_id": "s01",
      "scope": {
        "en": "test set",
        "zh": "测试集"
      }
    },
    {
      "facet": "task_format",
      "value": "open-qa",
      "basis": "inferred",
      "inferred_from": "card-content-description",
      "note": {
        "en": "QA pairs.",
        "zh": "问答对"
      }
    },
    {
      "facet": "specialty",
      "value": "multi-specialty",
      "basis": "inferred",
      "inferred_from": "card-content-description",
      "note": {
        "zh": "综合医疗问答，未限定专科",
        "en": "综合医疗问答; not limited to a specialty"
      }
    }
  ],
  "changelog": [
    {
      "date": "2026-11-12",
      "change": {
        "en": "Card converted from an earlier manual catalog; no new facts added; reclassified under taxonomy v2.",
        "zh": "由早期人工整理的目录转换生成，未新增事实；按分类体系 v2 重新归类。"
      }
    },
    {
      "date": "2026-11-12",
      "change": {
        "en": "License status split into license_status (what the license says) and hosting (what Mdexa does); the earlier conclusion is kept for traceability.",
        "zh": "把许可证状态拆成 license_status（许可证怎么写）与 hosting（本站怎么处理）；原结论保留以便追溯。"
      }
    },
    {
      "date": "2026-11-12",
      "change": {
        "en": "Added verbatim license excerpts and snapshot ids from captured page snapshots; no conclusion changed.",
        "zh": "补入来自已抓取页面快照的许可证逐字摘录与快照编号；未改动任何结论。"
      }
    },
    {
      "date": "2026-11-13",
      "change": {
        "en": "Added dataset-level tags: 5 (3 stated by the authors with excerpts, 2 inferred from the content). No conclusion changed.",
        "zh": "新增数据集标签 5 个（作者声明 3 个，附逐字摘录；据内容推断 2 个）；未改动任何结论。"
      }
    },
    {
      "date": "2026-11-13",
      "change": {
        "en": "Removed the stored size-tier tag; the tier is now computed from the size field when pages are built.",
        "zh": "移除存储的规模档位标签；档位改为生成页面时由 size 字段计算。"
      }
    },
    {
      "date": "2026-11-13",
      "change": {
        "en": "Added specialty tags: multi-specialty(inferred).",
        "zh": "新增专科标签：multi-specialty(inferred)。"
      }
    }
  ],
  "access_recipe": {
    "verified_on": "2026-11-14",
    "platform": "github",
    "repo_id": "FreedomIntelligence/Huatuo-26M",
    "source": {
      "api_url": "https://api.github.com/repos/FreedomIntelligence/Huatuo-26M",
      "fetched_on": "2026-11-14"
    },
    "access_state": "open",
    "files": [],
    "files_truncated": false,
    "commands": [
      {
        "command": "git clone https://github.com/FreedomIntelligence/Huatuo-26M",
        "evidence": {
          "excerpt": "git clone https://github.com/YOUR-USERNAME/YOUR-REPOSITORY",
          "doc": "github-clone",
          "url": "https://docs.github.com/en/repositories/creating-and-managing-repositories/cloning-a-repository",
          "fetched_on": "2026-11-14"
        }
      }
    ],
    "registration": null,
    "version": {
      "commit": null,
      "last_modified": "2024-03-14T04:54:23Z",
      "default_branch": "main"
    },
    "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": [
      "the platform publishes no checksum for these files; none was computed (we do not download data files)"
    ]
  },
  "_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: CMtMedQA (https://mdexa.com/datasets/cmtmedqa/); MedMCQA (https://mdexa.com/datasets/medmcqa/); CMB (Chinese Medical Benchmark) (https://mdexa.com/datasets/cmb/).",
    "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": "CMtMedQA",
        "url": "https://mdexa.com/datasets/cmtmedqa/",
        "json": "https://mdexa.com/datasets/cmtmedqa.json"
      },
      {
        "name": "MedMCQA",
        "url": "https://mdexa.com/datasets/medmcqa/",
        "json": "https://mdexa.com/datasets/medmcqa.json"
      },
      {
        "name": "CMB (Chinese Medical Benchmark)",
        "url": "https://mdexa.com/datasets/cmb/",
        "json": "https://mdexa.com/datasets/cmb.json"
      }
    ]
  }
}