EpistemeAI/Reasoning-Medical-27B

VISIONConcurrent Unit Cost:2Model Size:27BQuant:FP8Context Size:32kTool Calling:SupportedPublished:Jun 16, 2026License:apache-2.0Architecture:Transformer0.0K Open Weights Featherless Exclusive Cold

EpistemeAI/Reasoning-Medical-27B is a 27 billion parameter causal language model with a vision encoder, developed by EpistemeAI. Fine-tuned on 370,000 high-quality medical question-and-answer examples, it is designed for advanced medical reasoning across professional medicine, medical genetics, and clinical knowledge. The model natively supports a 262,144 token context length, extensible up to 1,010,000 tokens, and achieves 93.00% MedQA accuracy.

Loading preview...

Overview

EpistemeAI/Reasoning-Medical-27B is a 27 billion parameter causal language model with a vision encoder, developed by EpistemeAI. It was fine-tuned on a large-scale dataset of 370,000 high-quality medical question-and-answer examples, incorporating Chain-of-Thought reasoning to enhance step-by-step medical problem-solving. The model supports a native context length of 262,144 tokens, which can be extended up to 1,010,000 tokens using RoPE scaling techniques like YaRN.

Key Capabilities

  • Advanced Medical Reasoning: Designed for professional medicine, medical genetics, college biology/medicine, and clinical knowledge.
  • Multimodal Input: Supports text, image, and video inputs, enabling comprehensive analysis of medical cases.
  • High MedQA Accuracy: Achieves 93.00% accuracy on the MedQA benchmark in a 2-shot setup, outperforming several other models in its class.
  • HealthBench Professional Performance: Ranks highly on the HealthBench Professional benchmark with a score of 0.560.
  • Agentic Usage: Excels in tool calling capabilities and is compatible with frameworks like Qwen-Agent for building agent applications.
  • Thinking Mode: Operates in a default "thinking mode" to generate step-by-step reasoning, which can be preserved across messages for enhanced consistency in agent scenarios.

Good For

  • Medical Research and Information: Ideal for researchers and professionals seeking advanced medical reasoning and knowledge retrieval.
  • Educational Applications: Suitable for college-level biology and medicine education.
  • Multimodal Medical Analysis: Use cases involving diagnostic imaging (X-rays), medical charts, or video-based clinical observations.
  • Developing AI Agents: Its strong tool-calling and thinking preservation features make it suitable for building sophisticated medical AI agents.