Liquid AI Says LFM3-1.7B Beats Its 2.6B Model with 35% Fewer Parameters
Summary
Liquid AI releases LFM3, a family of six roughly 2-billion-parameter models spanning text, vision, audio, omni, and decision tasks, built on a 1.7B base. The company says LFM3-1.7B outperforms its LFM2.5-2.6B while using 35% fewer parameters. The open-weight models are available on Hugging Face and Liquid's API, with a 256K-token context window for text.
Key Points
- Liquid AI says LFM3-1.7B-Base is pre-trained on 22T tokens.
- Liquid AI reports the text model fits in about 1 GB at 4-bit precision.
- Liquid AI measures LFM3-1.7B decoding 1.8 to 2.8 times faster than the next model across three devices.