Responsible Use Guide for Llama

Responsibility

Responsible Use Guide: your resource for building responsibly

The Responsible Use Guide is a resource for developers that provides best practices and considerations for building products powered by large language models (LLM) in a responsible manner, covering various stages of development from inception to deployment. The resource goes into important safety considerations and provides developers with information on model and system level safeguards.

Responsible Use Guide

What's new

We’re excited to develop and release the latest Llama 3 models, the next generation of Llama which, like previous Llama versions, is licensed for commercial use. Llama 3 includes a collection of pretrained and instruction tuned generative models in 8B, 70B, and 405B parameter sizes to support text capabilities.

As part of the Llama update, we are updating our Responsible Use Guide to provide guidance on how to implement more advanced LLM capabilities and how to responsibly deploy these capabilities. Updates include:

Detailed overview of Llama 3 model
Additional system-level safety alignment and best practices
Responsibility considerations reference tool, multilinguality, and coding LLM use cases

Release resources

For the safe deployment of our Llama models, Meta recommends following the Responsible Use Guide and using the resource guide below which provides resources to facilitate the implementation of best practices, mapped to each stage of LLM product development.

Responsible use guide resources

Responsibility

Responsible Use Guide

To promote a responsible, collaborative AI innovation ecosystem, we’ve established a range of resources for all who use Llama 3: individuals, creators, developers, researchers, academics, and businesses of any size. The Responsible Use Guide is a resource for developers that provides recommended best practices and considerations for building products powered by LLMs.

Responsible Use Guide

Open approach

Our investment in open trust and safety

With Meta Llama Guard we are furthering our commitment to an open approach to build generative AI with trust and safety in mind. We believe in transparency, open science, and cross-collaboration, and to date, we’ve released over a thousand open-source libraries, models, datasets, and more.

Learn more

Research

Llama 3 Research Paper

For the Llama 3 launch, we are providing the research community with detailed insight into our safety testing approach and mitigations applied throughout out the model development lifecycle. The research paper will highlight the benchmarks and evaluations tested to ensure our model’s helpfulness and safety.