Medical datasets for fine-tuning that have a permissive license
13 of the 20 medical datasets for fine-tuning in the Mdexa database have a permissive license: Chinese Herbal Medicine Sentiment Analysis Dataset, Chinese-medical-dialogue-data（Toyhom）, CMB (Chinese Medical Benchmark), CMtMedQA, Huatuo-26M, HuatuoGPT-sft-data-v1, HuatuoGPT2-SFT-GPT4-140K, 医学考试题库 JSON 数据集, medical-o1-reasoning-SFT（中文子集）, Medical-Reasoning-SFT-Mega, MedMCQA, PubMedCausal, tmmlu++.
11 of them carry a source risk flag; a license verdict does not cover source risks.
Datasets 13
Chinese Herbal Medicine Sentiment Analysis DatasetVerified by Mdexa PermissiveLink only —
License permissive: first-hand data-scope statement(s) allow commercial use and redistribution. Risk flag(s): scraped. Link only.
Traditional Chinese medicine · Annotated corpora · Scraped
Chinese-medical-dialogue-dataToyhomVerified by Mdexa PermissiveLink only 792.1K QA pairs
License is permissive, but source risk: source not stated, so source copyright cannot be judged; link only.
Clinical medicine · QA, dialogue and instruction data · Unknown provenance
CMBChinese Medical BenchmarkVerified by Mdexa PermissiveLink only 11.2K questions
License is permissive, but source risk: questions come from copyrighted exams; link only.
Clinical medicine · Question banks and benchmarks · Clinical cases · Exam-derived
CMtMedQAVerified by Mdexa PermissiveLink only ≈68K dialogues
License is permissive, but source risk: derived from real patient text; link only.
Clinical medicine · QA, dialogue and instruction data · Real patient text
Huatuo-26MVerified by Mdexa PermissiveLink only ≥26M QA pairs
License is permissive, but source risk: content scraped from websites whose terms are not stated; derived from real patient text; link only.
Clinical medicine · QA, dialogue and instruction data · Scraped · Real patient text
HuatuoGPT-sft-data-v1Verified by Mdexa PermissiveLink only 226K rows
License is permissive, but source risk: source not stated, so source copyright cannot be judged; link only.
Clinical medicine · QA, dialogue and instruction data · Unknown provenance
HuatuoGPT2-SFT-GPT4-140KVerified by Mdexa PermissiveLink only —
License permissive: first-hand data-scope statement(s) allow commercial use and redistribution. Risk flag(s): derived, Model-generated. Link only.
Clinical medicine · QA, dialogue and instruction data · Derived · Model-generated
医学考试题库 JSON 数据集Verified by Mdexa PermissiveLink only —
License permissive: first-hand data-scope statement(s) allow commercial use and redistribution. Risk flag(s): Exam-derived. Link only.
Clinical medicine · Pharmacy and drugs · Chinese materia medica · Medical imaging · Public health · Question banks and benchmarks · Exam-derived
medical-o1-reasoning-SFT中文子集Verified by Mdexa PermissiveLink only ≈20.2K items
License is permissive, but source risk: generated by a model; the model's output terms need checking; link only.
Clinical medicine · QA, dialogue and instruction data · Model-generated
Medical-Reasoning-SFT-MegaVerified by Mdexa PermissiveLink only —
License permissive: first-hand data-scope statement(s) allow commercial use and redistribution. Risk flag(s): Model-generated. Link only.
Clinical medicine · QA, dialogue and instruction data · Model-generated
MedMCQAVerified by Mdexa PermissiveLink only —
License permissive: first-hand data-scope statement(s) allow commercial use and redistribution. Risk flag(s): Exam-derived, scraped. Link only.
Clinical medicine · Question banks and benchmarks · QA, dialogue and instruction data · Exam-derived · Scraped
PubMedCausalVerified by Mdexa PermissiveLink only —
License permissive: first-hand data-scope statement(s) allow commercial use and redistribution. Link only.
Biomedical literature · Annotated corpora · Literature
tmmlu++Verified by Mdexa PermissiveLink only —
License permissive: first-hand data-scope statement(s) allow commercial use and redistribution. Link only.
Traditional Chinese medicine · Pharmacy and drugs · Clinical medicine · Question banks and benchmarks · QA, dialogue and instruction data
Evidence
- Chinese Herbal Medicine Sentiment Analysis DatasetHugging Face card metadata · MIT · 2026-11-12
license: mit
- Chinese-medical-dialogue-data (Toyhom) Repository sidebar · MIT · 2026-11-12
MIT License Copyright (c) 2019 Toyhom
- CMB (Chinese Medical Benchmark)Repository license file · Apache-2.0 · 2026-11-12
Apache License Version 2.0, January 2004
- CMtMedQAHugging Face card metadata · MIT · 2026-11-12
license: mit
- Huatuo-26MREADME · Apache-2.0 · 2026-11-12
The Huatuo-26M dataset is licensed under Apache 2.0.
- HuatuoGPT-sft-data-v1Hugging Face card metadata · Apache-2.0 · 2026-11-12
license: apache-2.0
- HuatuoGPT2-SFT-GPT4-140KHugging Face card metadata · Apache-2.0 · 2026-11-12
license: apache-2.0
- 医学考试题库 JSON 数据集Hugging Face card metadata · Apache-2.0 · 2026-11-12
license: apache-2.0
- medical-o1-reasoning-SFT（中文子集）Hugging Face card metadata · Apache-2.0 · 2026-11-12
license: apache-2.0
- Medical-Reasoning-SFT-MegaHugging Face card metadata · Apache-2.0 · 2026-11-12
license: apache-2.0
- MedMCQAHugging Face card metadata · Apache-2.0 · 2026-11-12
"apache-2.0"
- PubMedCausalHugging Face card metadata · MIT · 2026-11-12
license: mit
- tmmlu++Hugging Face card metadata · MIT · 2026-11-12
license: mit
These are what the sources state; confirm the terms at the source before use.
Entries verified 2026-11-12 to 2026-11-14 · generated 2026-11-14 · Open in the database