---
icon: material/water-outline
---

# ![Refiners](/assets/logo_light.png)

<div style="text-align: center">
  <p>
    <strong>The simplest way to train and run <a href="/concepts/adapter/">adapters</a> on top of foundation models.</strong>
  </p>
  <a href="https://pypi.org/project/refiners/"><img alt="PyPI - Python Version" src="https://img.shields.io/pypi/pyversions/refiners"></a>
  <a href="https://badge.fury.io/py/refiners"><img alt="PyPI Status" src="https://badge.fury.io/py/refiners.svg"></a>
  <a href="https://github.com/finegrain-ai/refiners/blob/main/LICENSE"><img alt="license" src="https://img.shields.io/badge/license-MIT-blue"></a>
  <a href="https://finegrain.ai/bounties"><img alt="code bounties" src="https://img.shields.io/badge/code-bounties-blue"></a>
  <a href="https://discord.gg/mCmjNUVV7d"><img alt="chat" src="https://img.shields.io/discord/1179456777406922913?logo=discord&amp;logoColor=white&amp;color=%235765F2"></a>
</div>

Refiners is a microframework built on top of PyTorch with first-class citizen APIs for foundation model adaptation.

It is Open Source and published under the MIT License.