Models · Updated 28 Oct, 03:11 am IST
Ollama adds Liquid AI's LFM3 and d1 2B in v0.40.12
Why it matters for readers: You can try multimodal AI (text, image, audio) locally without large cloud models.
- LFM3 is a new family of small models from Liquid AI, including lfm3 (1.7B, 256K-token context), lfm3-vl, lfm3-audio, and lfm3-omni.1
- Liquid AI's d1 decision models are now available in a 2B size and include d1-omni for audio support.1
- The release adds support for running lfm3 and d1 2B on Apple hardware, with Linux and Windows support coming soon.1
- v0.40.12 also brings faster prompt processing on Apple silicon for Gemma 4, Flint, and Laguna models and other tooling fixes.1
Get a brief like this every morning
Uzha reads hundreds of sources and gives you the stories that matter for your work, with every source linked. Free.
Get started