---
title: README
emoji: 📈
colorFrom: indigo
colorTo: yellow
sdk: static
pinned: false
thumbnail: >-
  https://cdn-uploads.huggingface.co/production/uploads/6668520e95915a3f778f24ad/BVDhJNcPxSrXOgHMd1cNK.png
---
## About Us.

Meet the Open Meditron team. A collaborative project led by the Laboratory for Intelligent Global Health & Humanitarian Response Technologies [LiGHT Lab](https://www.yale-light.org/), at EPFL [School of Computer Science], Yale [School of Medicine], and Carnegie Mellon University Africa [School of Engineering].
<p align="center">
      <img src="https://cdn-uploads.huggingface.co/production/uploads/6668520e95915a3f778f24ad/BVDhJNcPxSrXOgHMd1cNK.png" alt="Meditron Logo" width = "500"/>
</p>

Building on our previous work, [MEDITRON](https://www.yale-light.org/post/meet-meditron), we present **Meditron-3** an open suite of large language models for medicine that has been co-designed and evaluated by a global group of mostly volunteer  clinicians, humanitarian actors, and data scientists. 
Developed by continuous pretraining on Llama3.1 using a new carefully curated corpus of open-source medical literature and clinical practice guidelines.

> The weights for **Meditron-3[70B]** and **Meditron-3[8B]** are now available on Huggingface.

#### 🌐 Stay connected with us on LinkedIn:

- **💼 Project leads:** [Annie Hartley](https://www.linkedin.com/in/annie-hartley-324832210/), [Alexandre Sallinen](https://www.linkedin.com/in/alexandre-sallinen-033359294/)