Medical-Specialized Reasoning-Medical-27B Released

·2026.07.28 16:19

Key point

Reasoning-Medical-27B, a model built on Qwen3.6-27B that maximizes medical reasoning ability, has been released.

Details

Reasoning-Medical-27B, a model designed for professional medicine, medical genetics, undergraduate biology, and clinical knowledge, has been released.

This model has the following technical features:

Dataset: Large-scale fine-tuning performed using 370,000 high-quality Q&A cases
Reasoning method: Chain-of-Thought (CoT) reasoning technique applied to improve step-by-step answering ability
Training optimization: Efficient training conducted using the GRPO trainer and Unsloth optimization methods

Model weights and a demo for this model are available via Hugging Face.

qwen unsloth reasoning-model fine-tuning grpo medical-llm

Read originalShare on Threads

This summary was generated automatically by AI. Check the original for the author's claims and context. Copyright belongs to the original author.

Our guide explains how the AI works. Report summary errors, attribution issues, or removal requests via Contact.