Liquid AI | Device-native foundation models.
Device-native foundation models.
Advanced intelligence for processors outside of data centers. Built for the latency, privacy, and hardware constraints of the physical world.
Models
- 0 LFMs shipped
- 0.0M downloads
- 0+ variants
Deployments
Phones, laptops, cars, space, e-commerce, financial services, bio, defense
Runtimes
llama.cpp, MLX, ONNX, CoreML, SGLang, vLLM
& more
Explore our models (LFMs)Explore Liquid Foundation Models
Models
LFM2.5-VL-DSpark: Accelerating vision-language models on edge and beyond
“Liquid AI's models are the best for the scaled production deployment - pareto-optimal, beating much larger rivals. That's why we use them at Shopify.”
Mikhail Parakhin CTO, Shopify
“Because of their very small footprint relative to their power, Liquid AI’s models are [...] an ideal candidate to run for yourself. If you are an enterprise, a small company, or a physical AI company, you can run these models on your own servers and still get tremendous power.”
Ion Stoica Co-Founder of Arena, Databricks, and Anyscale
“You’re creating [a new] class of where AI can be deployed and how it can be deployed. And I particularly love it because you’re rethinking how to optimize [and] how AMD as a chip developer can work with you as a practitioner of the state-of-the-art AI to create this whole new class of AI computation.”
Mark Papermaster CTO, AMD
“By advancing on-device speech, language understanding and reasoning with Liquid AI, we’re laying the foundation for the next generation of intuitive and multimodal in-car experiences.”
Jörg Burzer CTO, Mercedes-Benz
Achieve peak performance by fine-tuning LFMs for your use case.
The power of model customization directly in your hands. LFMs are designed for rapid customization to achieve peak performance for specified use cases at a footprint small enough to run locally on your chosen hardware. Our full-stack solution includes architecture, optimization, and deployment engines to accelerate the path from prototype to product.