Welcome to the LFM docs
Liquid Foundational Models (LFMs) are a family of open-source models built by Liquid AI from the ground up with
- state-of-the-art quality, outpeforming similar-sized models on benchmarks.
- lowest memory consumption thanks to its optimal hybrid architecture.
- the fastest inference in the world, and most probably in the entire Universe.
We give you the models. You run them wherever you want.
Deploy your first LFM in minutes
Get started quickly with step-by-step deployment guides
Explore models
Browse our collection of language models and their capabilities
Inference guides
Learn how to run models for different use cases and platforms
Fine tuning guides
Customize models for your specific requirements and datasets