EPFLiGHT/Meditron3-Qwen2.5-14B
EPFLiGHT/Meditron3-Qwen2.5-14B is a 14.8 billion parameter large language model developed by the OpenMeditron initiative, specialized in clinical medicine. Built upon the Qwen2.5-14B base model, it emphasizes equitable representation, contextual diversity, and evidence-based guidelines, particularly for limited-resource and humanitarian settings. This research-only model is designed to enhance clinical decision-making and access to medical information, trained on a diverse dataset including clinical guidelines and peer-reviewed medical publications. It features a 32768-token context length and is intended for research in computational linguistics and medicine.
Loading preview...
Meditron-3-Qwen2.5-14B: A Medical LLM for Research
EPFLiGHT/Meditron3-Qwen2.5-14B is a 14.8 billion parameter large language model developed by the OpenMeditron initiative, specifically designed for clinical medicine. It is built on the Qwen2.5-14B base model and focuses on equitable representation, contextual diversity, and real-world evidence-based guidelines, with a particular emphasis on limited-resource and humanitarian settings, neglected populations, and diseases.
Key Capabilities & Features
- Medical Specialization: Trained extensively on a unique dataset comprising clinical guidelines, peer-reviewed medical publications, synthetic differential diagnoses, and LLM-enhanced medical MCQs.
- Research Focus: Primarily intended as a research tool to study and evaluate the potential of LLMs in clinical decision-making and providing access to medical information.
- Foundation Model: Released as a static foundation model, not instruction-tuned, but suitable for adaptation to specific downstream tasks via techniques like RLHF or DPO.
- Context Length: Supports a substantial context window of 32768 tokens.
- Evaluation: Shows competitive performance on medical multiple-choice question benchmarks like MedmcQA and MedQA compared to the base Qwen2.5-14B-Instruct model.
When to Consider Using This Model
- Medical AI Research: Ideal for researchers exploring LLM applications in clinical medicine, especially concerning global health equity and diverse medical contexts.
- Downstream Task Adaptation: Suitable as a base model for fine-tuning on specific medical question-answering tasks or other clinical applications.
- Studying Medical Information Access: Useful for investigating how LLMs can enhance access to evidence-based medical information.
Important Note: This model is research-only and not validated for direct medical use or clinical decision-making. It should not replace professional medical advice.