FreeLLMLiquid AI launches LFM3 series with six 2B‑class multimodal models
Liquid AI introduced LFM3, a set of six open‑weight 2 billion‑parameter models covering text, vision, audio, omni and decision tasks.
Liquid AI announced the LFM3 series on 27 October 2026. The lineup includes six open‑weight models, each built on a 2 billion‑parameter scale.
The models span multiple modalities: text, vision, audio, omni‑modal and decision‑making capabilities. All are released under an open‑weight license.
LFM3‑1.7B features a tri‑hybrid backbone and supports a 256 K token context window. It can be stored in roughly 1 GB when quantised to 4‑bit precision.
The release follows a trend of making multimodal AI more accessible for research and development, offering larger context windows than many earlier 2 B‑class models.
What it means for you
For developers and researchers, LFM3 provides free, open‑weight multimodal models that can run on modest hardware, expanding options beyond text‑only systems. Users seeking local or offline AI can now experiment with vision and audio capabilities without paying for commercial services.
Sources
- Liquid AI Releases LFM3: 6 Open-Weight 2B-Class Multimodal Models, MarkTechPost, 27 Oct 2026
Written automatically from the sources above, then checked by our code: every number and name must appear in them, and no sentence may copy a source. Spotted a mistake? Report it.
Which AI can you use for free today?
Every free AI model, ranked by blind human votes, with the memory and speed it needs on your own GPU.