• Models
  • Research
  • Engineering Blog
Connect

Building the fastest, most compute-efficient, and capable foundation models so intelligence can live anywhere.

Est. 2023

Models

  • Liquid Foundation Models (LFMs)
  • Model Playground
  • Hugging Face
  • Apollo

Developer Resources

  • Liquid Edge AI Platform (LEAP)
  • Demos
  • Documentation
  • Hackathons
  • Discord
  • GitHub

Industry Solutions

  • Automotive
  • Consumer electronics
  • E-commerce
  • Financial services
  • Healthcare
  • Industrial
  • Defense

Research

  • Liquid Labs
  • Automated Foundation Model Design
  • Blog

Company

  • About
  • Careers
  • News
  • Events
  • FAQ
  • Contact
  • Shoutouts

© 2026 Liquid AI, Inc.

  • Terms
  • Privacy
  • Press

News/All

News and Updates

AllModelsEngineeringCase studiesCompany newsResearch
Tokenizer Expansion: Upgrading a Model's Tokenizer in Place
JUL 21, 2026Engineering

Tokenizer Expansion: Upgrading a Model's Tokenizer in Place

07.07
07.07Engineering

Reducing Doom Loops with Final Token Preference Optimization

JUL 7, 2026Engineering

Reducing Doom Loops with Final Token Preference Optimization

06.30
06.30Engineering

IFStruct: Measuring structured-output compliance

JUN 30, 2026Engineering

IFStruct: Measuring structured-output compliance

06.25
06.25Models

LFM2.5-230M: Built to Run Anywhere

JUN 25, 2026Models

LFM2.5-230M: Built to Run Anywhere

  • 07.07.2026Engineering
    07.07.2026EngineeringReducing Doom Loops with Final Token Preference Optimization
  • 06.30.2026Engineering
    06.30.2026EngineeringIFStruct: Measuring structured-output compliance
  • 06.25.2026Models
    06.25.2026ModelsLFM2.5-230M: Built to Run Anywhere
  • 06.18.2026Models
    06.18.2026ModelsLFM2.5 Retrievers: Bi-directional LFMs for Fast Multilingual Search
  • 06.16.2026Publications
    06.16.2026PublicationsFixed-Point Reasoners: Stable and Adaptive Deep Looped Transformers
  • 05.31.2026Publications
    05.31.2026PublicationsNeural Low-Discrepancy Sequences
  • 05.29.2026Publications
    05.29.2026PublicationsSwitchCraft: A Programmatic Framework for Designing State-Switching Proteins
  • 05.28.2026Models
    05.28.2026ModelsLFM2.5-8B-A1B: An Even Better On-Device Mixture of Experts
  • 05.20.2026Publications
    05.20.2026PublicationsMeMo: Model as a Memory
  • 04.23.2026Press
    04.23.2026PressLiquid AI and Mercedes-Benz partner to scale embedded in-car intelligence
  • 04.22.2026Publications
    04.22.2026PublicationsPreconditioned DeltaNet: Curvature-aware Sequence Modeling for Linear Recurrences
  • 04.08.2026Models
    04.08.2026ModelsLFM2.5-VL-450M: Structured Visual Intelligence, Edge to Cloud
  • 03.31.2026Models
    03.31.2026ModelsLFM2.5-350M: No Size Left Behind
  • 03.26.2026Publications
    03.26.2026PublicationsThe Key to State Reduction in Linear Attention: A Rank-based Perspective
  • 03.16.2026External article
    03.16.2026External articleWhy Bigger Isn’t Always Better in AI