README.md

---
title: README
colorFrom: gray
colorTo: gray
sdk: static
pinned: false
---

<p align="center">
  <a href="https://while.ai"><img src="https://huggingface.co/spaces/while-ai/README/resolve/main/banner-wai.png" alt="wai, the While whale. Models improve while they work." width="720"></a>
</p>

<p align="center"><code>MID-TRAINING AND POST-TRAINING FOR LANGUAGE MODELS</code></p>

**Models improve while they work.** A scientific RL and SFT post-training library.

While is the mid-training and post-training platform for language models. Open models, the latest SFT and RL algorithms, and an open-source SDK that builds and optimizes datasets from production traces. Every gain is measured on a held-out set.

Everything here was made with the SDK and published with the numbers behind it. A model ships when it beats its base on a held-out set with an interval that clears zero. If it does not, it does not ship.

```bash
pip install whileai
```

```python
import whileai as wai
```

wai is While's whale and the alias of the whileai SDK.

[Docs](https://docs.withwhile.com) | [Platform](https://while.ai) | [SDK](https://github.com/whilehq/whileai-sdk) | [Recipes](https://github.com/whilehq/whileai-sdk/tree/main/recipes)

Collections group this org by use case.