Home

Awesome

<div align="center">

Optimum-TPU

<h4>Take the most out of Google Cloud TPUs with the ease of 🤗 transformers</h4>

Documentation license Optimum TPU / Test TGI on TPU

</div>

Tensor Processing Units (TPU) are AI accelerator made by Google to optimize performance and cost from AI training to inference.

This repository exposes an interface similar to what Hugging Face transformers library provides to interact with a magnitude of models developed by research labs, institutions and the community.

We aim at providing our user the best possible performances targeting Google Cloud TPUs for both training and inference working closely with Google and Google Cloud to make this a reality.

Supported Model and Tasks

We currently support a few LLM models targeting text generation scenarios:

Installation

optimum-tpu comes with an handy PyPi released package compatible with your classical python dependency management tool.

pip install optimum-tpu -f https://storage.googleapis.com/libtpu-releases/index.html

export PJRT_DEVICE=TPU

Inference

optimum-tpu provides a set of dedicated tools and integrations in order to leverage Cloud TPUs for inference, especially on the latest TPU version v5e.

Other TPU versions will be supported along the way.

Text-Generation-Inference

As part of the integration, we do support a text-generation-inference (TGI) backend allowing to deploy and serve incoming HTTP requests and execute them on Cloud TPUs.

Please see the TGI specific documentation on how to get started.

JetStream Pytorch Engine

optimum-tpu provides an optional support of JetStream Pytorch engine inside of TGI. This support can be installed using the dedicated CLI command:

optimum-tpu install-jetstream-pytorch

To enable the support, export the environment variable JETSTREAM_PT=1.

Training

Fine-tuning is supported and tested on the TPU v5e. We have tested so far:

You can check the examples: