|
--- |
|
title: README |
|
emoji: 📉 |
|
colorFrom: red |
|
colorTo: red |
|
sdk: static |
|
pinned: false |
|
--- |
|
|
|
The Nanotron team focus on sharing open knowledge and developping open-source libraries for efficient distributed training of large-scale AI models. |
|
|
|
Some of its contributions are: |
|
|
|
- the [Nanotron library](https://github.com/huggingface/nanotron) |
|
- the [Picotron library](https://github.com/huggingface/picotron) |
|
- the [Ultrascale-Playbook](https://huggingface.co/spaces/nanotron/ultrascale-playbook), a comprehensive book covering all distributed/parallelisation and low-level techniques that can be used to efficiently train models at the largest scales. |
|
|
