LFM2 (Liquid Foundation Models 2)
modelEfficient foundation models using a hybrid backbone combining gated short convolutions with grouped query attention. 350M to 24B parameters (MoE, 2.3B active). Optimized for edge and on-device deployment.
LFM2-24B achieves 239 tok/s (#3 speed ranking on AA). Also includes vision-language (LFM2.5-VL), speech, and pharmaceutical variants (with Insilico Medicine). AA Intelligence Index: 10 (24B). Apache 2.0.
Model Details
Architecture MOE
Parameters 24B
Active params 2.3B
Variants
| Name | Parameters | Notes |
|---|---|---|
| LFM2-350M | 350M | — |
| LFM2-1.2B | 1.2B | — |
| LFM2-2.6B | 2.6B | — |
| LFM2-8B-A1B | 8.3B | — |
| LFM2-24B-A2B | 23.8B | — |
Paper
arXiv: 2511.23404