Skip to content

Dataset · FreedomIntelligence

Medical SFT reasoning from DeepSeek-R1

FreedomIntelligence/Medical-R1-Distill-Data

Medical tasks with R1 reasoning chains for model fine-tuning.

What's inside

An SFT dataset distilled from DeepSeek-R1 (Full Power) on verifiable medical tasks from HuatuoGPT-o1. Contains text question-answer pairs with reasoning; languages are English and Chinese, size ranges from 10k to 100k examples. Requests to DeepSeek-R1 were made via the native API; a separate Chinese version exists.

What you can do with it

  • Fine-tuning medical LLMs to reason
  • Initializing a model with R1 chains
  • Evaluating QA on medical tasks
Tasks
Question answering, Text generation
Languages
English, Chinese
Size
10K–100K rows
Downloads this month
2.7K
Likes
78

Models trained on it

How to load it

from datasets import load_dataset

ds = load_dataset("FreedomIntelligence/Medical-R1-Distill-Data")