Note

This is only a preview/placeholder for LMQL’s documentation. The full documentation will be available on official launch.

Welcome to LMQL!

LMQL (Language Model Query Language) is a programming language for large language model interaction. It facilitates LLM interaction by combining the benefits of natural language prompting with the expressiveness of Python. With only a few lines of LMQL code, users can express advanced, multi-part and tool-augmented LM queries, which then are optimized by the LMQL runtime to run efficiently as part of the LM decoding loop.

LMQL is a research project by the Secure, Reliable, and Intelligent Systems Lab at ETH Zürich.

Quick Start

To get started locally, check out the Getting Started section. For smaller experiments, you can also use the web-based LMQL Playground:

Explore LMQL

Playground IDE GitHub Repo

Run Locally

pip install lmql

To run LMQL locally, read the Getting Started section of the documentation.

Contents

• Getting Started
  • Running LMQL Programs
• LMQL Language
  • Scripted Prompting
• Python Interoperability
  • Python Integration
  • 🦜🔗 LangChain Integration
• Setting Up a Development Environment
  • Development without GPU