Liquid AI dropped an experimental 3B model that punches way above its weight
Liquid AI just rolled out LFM2-2.6B-Exp, an experimental checkpoint built entirely with pure reinforcement learning. Despite its 2.6B size, it outperforms competing 3B models across instruction, knowledge, and math benchmarks, even topping DeepSeek R1-0528 on IFBench despite being 263 times smaller. The weights are already live on Hugging Face.