EpistemeAI/Reasoning-Medical0.1-E4B-sft
EpistemeAI/Reasoning-Medical0.1-E4B-sft is a supervised fine-tuned medical reasoning model based on unsloth/gemma-4-E4B-it, developed by EpistemeAI. It was trained on approximately 100,000 medical reasoning examples to enhance structured medical explanation, biomedical question answering, and differential reasoning. This model is specifically designed for advanced medical reasoning in professional medicine, medical genetics, college biology/medicine, and clinical knowledge.
Loading preview...
EpistemeAI/Reasoning-Medical0.1-E4B-sft: Specialized Medical Reasoning
This model, developed by EpistemeAI, is a supervised fine-tuned (SFT) variant of unsloth/gemma-4-E4B-it, specifically engineered for advanced medical reasoning. It was trained on a large-scale dataset of approximately 100,000 curated medical reasoning records, incorporating Chain-of-Thought reasoning to improve step-by-step medical question analysis.
Key Capabilities
- Enhanced Medical Reasoning: Excels in structured medical explanation, biomedical question answering, and differential reasoning.
- Educational Focus: Designed for medical education, research support, and clinical knowledge exploration.
- Benchmark Performance: Achieves a 0.760 accuracy on MedQA (4 options), outperforming several other models in its class.
- Safety-Oriented: Achieved an 87.5/100 mean BioSafeBench score, demonstrating robust refusal and redirection for unsafe biological questions.
Intended Use Cases
- Medical education and study support
- Biomedical and clinical reasoning practice
- Medical multiple-choice question reasoning
- Literature review assistance and research hypothesis exploration
- Drafting clinician-reviewed explanations
Important Considerations
This model is intended as a reasoning assistant and not for autonomous diagnosis, treatment, or replacement of licensed medical professionals. Users must verify outputs against trusted medical references and use clinician review for any patient-facing or clinical workflows. It may produce incorrect or outdated information and should be used with caution, especially in healthcare contexts.