LFM3

Latest news

1. 28 Oct
   Ollama adds Liquid AI's LFM3 and d1 2B in v0.40.12

   Ollama's v0.40.12 release introduces Liquid AI's LFM3 family of small models (text, vision, audio, and omni variants) and 2B d1 decision models built on LFM3. The update adds Apple hardware support for lfm3 and d1 2B and includes speed and compatibility improvements across other models and tools.

   Models · 1 source

2. 27 Oct
   Hugging Face launches LFM3 multimodal 2B-class family for edge specialization

   Hugging Face announced LFM3, a new generation of Liquid Foundation Models offering text, vision, audio, omni and decision models on a shared backbone in the ~2B parameter range. The models are designed for higher capability per parameter, single-GPU fine-tuning, and on-device deployment with features like a 256K context window for the text model.

   Models · 1 source

   

Articles

No articles about LFM3 yet.