InstructLab

An open source project for enhancing LLMs used in generative AI applications

The project on GitHub Read our Documentation Read the paper
A cute dog

Periodic release cycle for models and data

InstructLab
Open source community experiments with skills
Contributors open pull requests for InstructLab
New skills merged into community LLM
New skills recipes
Approved pull requests
Updated community model

How it works

Taxonomy

A taxonomy of skills and knowledge helps to identify gaps in desired capabilities, and then generate enough diversity in synthetic data to tune base models efficiently.

$

Try out InstructLab

To get started, download the ilab command line interface and a quantized version of the latest model from GitHub.

You can experiment locally with synthetic data generation, model training, serving and chatting with the model to see if it produces the desired outputs.

Check out the project repos