LMQL

LMQL

LMQL is a programming language tailored for large language models (LLMs). It offers robust and modular LLM prompting through the use of types, templates, constraints, and an optimizing runtime. It simplifies the creation of complex prompts by allowing procedural programming techniques in a query-like syntax. Created by the SRI Lab at ETH Zurich, LMQL supports features such as nested queries, scripted prompting, and custom constraints. It also provides a Playground IDE for ease of use.

Features

  • Nested Queries
  • Scripted Prompting
  • Custom Constraints
  • Optimizing Runtime
  • Playground IDE
  • Local Model Support
  • Tool Augmentation
  • High-level Constraint Management

Use Cases

  • Developers
  • Researchers
  • Data Scientists
  • AI Practitioners
  • Educators
  • Hobbyists
  • Enterprises
  • AI Enthusiasts
Visit LMQL

May be interested in