EpistemeAI/Reasoning-Medical0.1-E4B-sft
EpistemeAI/Reasoning-Medical0.1-E4B-sft is a 7.9 billion parameter instruction-tuned causal language model developed by EpistemeAI, fine-tuned from unsloth/gemma-4-E4B-it. It is specifically optimized for advanced medical reasoning, biomedical question answering, and clinical education, trained on approximately 100,000 medical reasoning examples. This model excels at structured medical explanation and differential reasoning, making it suitable for research and educational support in medical domains.
Loading preview...
Overview
EpistemeAI/Reasoning-Medical0.1-E4B-sft is a 7.9 billion parameter supervised fine-tuned (SFT) medical reasoning model, built upon unsloth/gemma-4-E4B-it. Developed by EpistemeAI, it was trained on a curated dataset of approximately 100,000 medical reasoning examples, incorporating Chain-of-Thought techniques to enhance step-by-step medical problem-solving.
Key Capabilities
- Advanced Medical Reasoning: Excels in professional medicine, medical genetics, and college biology/medicine contexts.
- Biomedical Question Answering: Designed to provide structured explanations and answers to complex medical queries.
- Clinical Education Support: Aids in differential reasoning and educational analysis for medical students and professionals.
- Efficient Fine-tuning: Utilizes Unsloth optimization for faster training.
Intended Use Cases
This model is a reasoning assistant for:
- 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 Safety Notice
This model is for benign medical and scientific reasoning only. It is not intended for autonomous diagnosis, treatment, or any high-stakes medical decision-making without professional oversight. Users must verify outputs against trusted medical references and comply with all applicable legal, ethical, and safety standards. The model may produce incorrect or outdated information and should be used with caution.