Skip to content- Sources: primary, discussion, discussion
- Summary: LFM3-1.7B-Base has 24 layers mixing 13 gated short convolutions, 5 sliding-window attention layers and 6 global attention layers, and was pre-trained on 22T tokens. Six models branch from it, including LFM3-VL-2B with video input, LFM3-Audio-2B, LFM3-Omni-2B and two decision models. Liquid AI reports LFM3-1.7B scores 38.42 on WebArena-Verified against 3.02 for LFM2.5-2.6B, a vendor-reported result.
- Why it matters: Small open models with tool calling and long context make local agents and edge deployment practical without cloud inference, pending independent evaluation.