LFM3-1.7B

Use it to build and run local agents, work with long text contexts, and fine-tune a model for specialized tasks.

Availability

Available today on Hugging Face, Liquid AI Playground, and Liquid AI API

Pricing model

Tokens are free

Deployment

Can be fine-tuned locally and deployed where an application runs

Privacy

Not established

Licence

Open weights

• The base and post-trained models are available today on Hugging Face, Liquid AI Playground, and Liquid AI API. ^[1]

  The base and post-trained models are open-weight and available today on Hugging Face , our Playground , and the Liquid AI API .

• LFM3-1.7B is a text model for local agents and specialized applications. ^{[1] [2] [3]}

  LFM3-1.7B is a text model for local agents and specialized applications with a 256K-token context window.

• The text model has a 256K-token context window. ^{[1] [2] [3] [4]}

  LFM3-1.7B is a text model for local agents and specialized applications with a 256K-token context window.

• The model is post-trained to run in agentic harnesses such as Hermes Agent, Pi, and OpenClaw. ^{[1] [2] [3]}

  It is post-trained to run in agentic harnesses such as Hermes Agent, Pi, and OpenClaw.

• The models can be fine-tuned locally, including with a short fine-tuning run on a single GPU. ^{[1] [2] [3]}

  You can adapt it to your domains, harnesses, and tools with a short fine-tuning run on a single GPU.

• New models bring live speech translation and voice features Daily brief 2026-10-28