Gemma

Gemma is a family of open-weights Large Language Model (LLM) by Google DeepMind, based on Gemini research and technology.

This repository contain the implementation of the gemma PyPI package. A JAX library to use and fine-tune Gemma.

For examples and uses-cases, see our documentation. Please report issues and feedback in our GitHub.

Learn more about Gemma

• To use this library: https://gemma-llm.readthedocs.io/
• Technical reports for metrics and model capabilities:
  • Gemma 1
  • Gemma 2
• Other Gemma implementations and doc on the Gemma ecosystem: https://ai.google.dev/gemma/docs

Quick start

Installation

1. Install JAX for CPU, GPU or TPU. Follow instructions at the JAX website.

2. Run

   pip install gemma
   

Downloading the models

To download the model weights. See our documentation.

Examples

Our documentation contain various Colabs and tutorial for:

• Sampling
• Fine-tuning
• LoRA

Additionally, our examples/ folder contain additional scripts to fine-tune and sample with Gemma.

System Requirements

Gemma can run on a CPU, GPU and TPU. For GPU, we recommend a 8GB+ RAM on GPU for the 2B checkpoint and 24GB+ RAM on GPU for the 7B checkpoint.

This is not an official Google product.