# FreedomIntelligence releases HuatuoGPT-3-9B medical LLM with open usage guides

Digest AI · Generative AI & Models · published 2026-09-17T16:30:07Z

Canonical: https://digestai.news/story/freedomintelligence-releases-huatuogpt-3-9b-medical-llm-with-open-usage-guides

## Summary

FreedomIntelligence has made its new HuatuoGPT-3-9B model publicly available on Hugging Face. The 9‑billion‑parameter model is built on Qwen3.5‑9B and uses a One‑stage Policy Optimization (OnePO) technique that adapts a language model to medical tasks without a separate supervised fine‑tuning stage. The release includes the model weights, training code, a medical reinforcement‑learning dataset, and an 8‑billion‑parameter rubric grader for evaluation.

The post provides step‑by‑step instructions for deploying HuatuoGPT‑3‑9B with several popular inference stacks. Users can run the model via the Transformers library, set up an OpenAI‑compatible API with vLLM or SGLang, or launch a Docker Model Runner container. Example snippets show how to send image‑text prompts—such as asking “What animal is on the candy?”—and retrieve concise answers. The guide also notes that the model supports a “thinking mode” where it generates reasoning before the final answer, and that the model can handle up to 4,096 new tokens per request.

A citation to the underlying OnePO paper (Chen et al., 2026) is included, encouraging academic reuse and further research into domain‑specific policy optimization for large language models.

## Key points

- HuatuoGPT-3-9B is a 9‑billion‑parameter medical LLM built on Qwen3.5‑9B using One‑stage Policy Optimization.
- The model, training code, medical RL dataset, and an 8‑billion‑parameter rubric grader are released on Hugging Face.
- Usage guides cover Transformers pipelines, vLLM, SGLang, and Docker Model Runner for image‑text and pure text queries.

## Why it matters

An open‑source medical LLM with ready‑to‑use deployment scripts lowers barriers for developers and researchers building AI‑driven healthcare tools, accelerating innovation while highlighting responsible use concerns.

## Sources

1. [FreedomIntelligence/HuatuoGPT-3-9B · Hugging Face](https://huggingface.co/FreedomIntelligence/HuatuoGPT-3-9B) (huggingface.co, 2026-09-17, primary source)

Part of the developing story: [Open Source AI Infrastructure and Medical Model Expansion](https://digestai.news/thread/ggml-org-merges-hc-ops-into-qwen4exp-graph-in-llama-cpp) (4 stories)

## Cite

Digest AI, "FreedomIntelligence releases HuatuoGPT-3-9B medical LLM with open usage guides", 17 September 2026, https://digestai.news/story/freedomintelligence-releases-huatuogpt-3-9b-medical-llm-with-open-usage-guides

---

Written by Digest AI's editorial model from the linked sources; the sources are the record. Headlines, digests and key points are written by Digest AI and may be quoted with a link to the story page. Linked articles belong to their publishers. Terms: https://digestai.news/terms#reuse
JSON: https://digestai.news/story/freedomintelligence-releases-huatuogpt-3-9b-medical-llm-with-open-usage-guides.json