FreedomIntelligence/HuatuoGPT-3-27B · Hugging Face

from FreedomIntelligence:

HuatuoGPT-3-27B is a medical LLM built on Qwen3.8-27B with One-stage Policy Optimization (OnePO). OnePO adapts language models to medicine in a single reinforcement-learning stage, without preceding domain-specific supervised fine-tuning. Teacher responses provide temporary guidance and are retired as the model improves.

We release the training code, medical RL dataset, and 8B rubric grader.

(last week they released https://huggingface.co/FreedomIntelligence/HuatuoGPT-3-9B)