💬 Welcome to LMQL

LMQL (Language Model Query Language) is a programming language for large language model (LM) interaction. It facilitates LLM interaction by combining the benefits of natural language prompting with the expressiveness of Python. It has a focus on multi-part prompting and enables novel forms of LM interaction via scripting, constraint-guided decoding, tool augmentation, and efficiency.

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

Quick Start

To get started with LMQL, check out the 🚀 Getting Started guide.

We also provide the following resources to get started with LMQL:

Explore LMQL

Playground IDE 🚀 Getting Started Guide 🖼️ Examples Gallery

Run Locally

pip install lmql

To run LMQL locally, read the Installation Instructions section of the documentation.

Community

💬 Discord Server 🐥 Twitter ✉️ E-Mail

Contents

• 🚀 Getting Started
• 💾 Installation Instructions
• ⚡️ Playground IDE

📖 LMQL Language

• Overview
• Scripted Prompting
• Constraints
• Decoders
• Models
• External Functions and Tools

🔗 Python Integration

• LMQL in Python
• LangChain Integration 🦜🔗
• LlamaIndex Integration 🦙
• LMQL and Pandas 🐼
• Output Streaming
• Comparison with Other Libraries

💬 Contribute

• Setting Up a Development Environment
• LMQL in Docker
• Discord
• GitHub Issues