FreedomIntelligence/HuatuoGPT-3-8B

Hugging Face
TEXT GENERATIONConcurrency Cost:1Model Size:8BQuant:FP8Ctx Length:32kPublished:Mar 20, 2026License:apache-2.0Architecture:Transformer0.0K Open Weights Warm

HuatuoGPT-3-8B is an 8 billion parameter medical large language model developed by FreedomIntelligence, built on the Qwen3-8B-Base backbone. It is specifically fine-tuned for medical expertise using the SeedRL paradigm, an RL-only domain adaptation method. This model excels at medical question-answering and reasoning, providing detailed responses with an explicit thinking process.

Loading preview...

Overview

HuatuoGPT-3-8B is an 8 billion parameter medical large language model developed by FreedomIntelligence. It is built upon the Qwen3-8B-Base backbone and has been specialized for medical applications through a unique training approach called SeedRL. This method involves an RL-only domain adaptation paradigm, transforming a general-purpose base model into a medical expert in a single reinforcement learning stage.

Key Capabilities

  • Medical Expertise: Specifically designed and optimized for medical domain knowledge and tasks.
  • Reasoning Block: By default, the model operates in a "thinking mode," providing a <think>...</think> reasoning block before its final response, which can be valuable for understanding its decision-making process.
  • Efficient Domain Adaptation: Utilizes the SeedRL paradigm for effective and streamlined adaptation to the medical domain.

Usage Notes

  • The model can be used similarly to Qwen3-8B and is compatible with deployment tools like vLLM and SGLang.
  • Its default "thinking mode" output format includes an explicit reasoning trace, which is a notable feature for interpretability in medical contexts.