# Tri-hybrid architecture

Liquid AI released the LFM3 family, which puts text, vision, audio, omni and decision models on one shared 2B-class backbone, all open-weight and fine-tunable on a single GPU. Against the previous LFM2 family, LFM3-1.7B outperforms LFM2.5-2.6B with 35% fewer parameters, scoring 38.42 versus 3.02 on WebArena-Verified. The text model fits in about 1 GB at 4-bit precision and supports a 256K context window for on-device agents.

Liquid AI released the LFM3 family, which puts text, vision, audio, omni and decision models on one shared 2B-class backbone, all open-weight and fine-tunable on a single GPU. Against the previous LFM2 family, LFM3-1.7B outperforms LFM2.5-2.6B with 35% fewer parameters, scoring 38.42 versus 3.02 on WebArena-Verified. The text model fits in about 1 GB at 4-bit precision and supports a 256K context window for on-device agents.