README.md

---
title: BioMistral
emoji: 👨‍⚕️
colorFrom: purple
colorTo: blue
sdk: static
pinned: true
---

<p align="center">
<!--   <img src="https://huggingface.co/BioMistral/BioMistral-7B/resolve/main/Image%2BTitle.png?download=true" alt="drawing" width="600"/> -->
  <img src="https://huggingface.co/BioMistral/BioMistral-7B/resolve/main/wordart_blue_m_rectangle.png?download=true" alt="drawing" width="250"/>
</p>

BioMistral 7B is the best performing open-source medical Large Language Model (LLM) in this weight category. 

We've publicly released the weights for BioMistral 7B on Huggingface and our full multilingual benchmark on GitHub.

- 📰 Paper: BioMistral: <a href="https://arxiv.org/abs/2402.10373"> A Collection of Open-Source Pretrained Large Language Models for Medical Domains</a> (pre-print)

- 📊 Multilingual medical benchmark : <a href="https://huggingface.co/datasets/BioMistral/BioInstructQA">BioMistral/BioInstructQA</a>

- 👩‍💻 GitHub: <a href="https://github.com/BioMistral/BioMistral">BioMistral/BioMistral</a>

This project is the result of the collaboration between:

<table>
<tbody>
  <tr>
    <td>🏛️ <a href="https://lia.univ-avignon.fr/">LIA - Avignon University (1)</a></td>
    <td>🏛️ <a href="https://www.ls2n.fr/">LS2N - Nantes University (3)</a></td>
  </tr>
  <tr>
    <td>🏥 <a href="https://www.chu-nantes.fr/unite-recherche-2">Nantes University Hospital (2)</a></td>
    <td>🏢 <a href="https://zenidoc.fr/">Zenidoc (4)</a></td>
  </tr>
</tbody>
</table>

**Authors** : Yanis LABRAK (1,4) ; Adrien BAZOGE (2,3) ; Emmanuel MORIN (3) ; Pierre-Antoine GOURAUD (2) ; Mickaël ROUVIER (1) ; Richard DUFOUR (3)

**Caution:** We recommend using BioMistral 7B strictly as a research tool and advise against deploying it in production environments for natural language generation or any professional health and medical purposes.