Skip to content
Not available in this workspace
OpenRouterOpenRouter
© 2026 OpenRouter, Inc

Product

  • Chat
  • Rankings
  • Benchmarks
  • Apps
  • Discover
  • Models
  • Collections
  • Providers
  • Pricing
  • Enterprise
  • Labs

Company

  • About
  • Blog
  • Careers
    Hiring
  • Privacy
  • Terms of Service
  • Support
  • Works With OR
  • Data

Developer

  • Documentation
  • API Reference
  • Developer Platform
  • Status

Connect

  • Discord
  • GitHub
  • LinkedIn
  • X
  • YouTube
Favicon for liquid

Liquid: LFM 7B

liquid/lfm-7b

LFM-7B, a new best-in-class language model. LFM-7B is designed for exceptional chat capabilities, including languages like Arabic and Japanese. Powered by the Liquid Foundation Model (LFM) architecture, it exhibits unique features like low memory footprint and fast inference speed.

LFM-7B is the world’s best-in-class multilingual language model in English, Arabic, and Japanese.

See the launch announcement(opens in new tab) for benchmarks and more info.

Modalities

Context

33K

Released

Jan 25, 2025

Knowledge Cutoff

Jun 2024

About Liquid: LFM 7B

OpenRouter makes Liquid: LFM 7B available through a unified, OpenAI-compatible API using the model ID liquid/lfm-7b.

Liquid: LFM 7B accepts text and returns text. It has a 32,768-token context window.

It was released on January 25, 2025; its knowledge cutoff is June 30, 2024.

More models from Liquid

  • LFM2.5-Embedding-350M (free)

Frequently asked questions

LFM-7B, a new best-in-class language model. LFM-7B is designed for exceptional chat capabilities, including languages like Arabic and Japanese. Powered by the Liquid Foundation Model (LFM) architecture, it exhibits unique features like low memory footprint and fast inference speed. LFM-7B is the world’s best-in-class multilingual language model in English, Arabic, and Japanese.

LFM 7B has a 32,768 token context window.

LFM 7B was released on January 25, 2025. Its knowledge cutoff is June 30, 2024.

ActivityFAQ

Activity

Token volume and request traffic to this model over time.