Tagged with

3 articles found

Liquid AI’s LFM2-24B: A 24B-Parameter MoE That Runs on 32GB RAM and Makes Cloud APIs Look Overpriced

Liquid AI’s new sparse MoE model activates only 2.3B parameters per token, delivering server-class AI performance on consumer hardware while challenging the cloud-only paradigm.

#lfm2#liquid-ai#moe

hybrid-architecture

LFM 2.5: The 1.2B Parameter Model That Makes Bigger Look Dumber

Liquid AI’s LFM 2.5 challenges everything we thought we knew about model scaling, delivering 3x performance in a package small enough for your phone

#hybrid-architecture#liquid-ai#small-language-models

lfm25

Liquid AI’s LFM2.5: The Tiny Model That Promises Everything But Delivers… Exactly What It Should

Liquid AI’s 1.2B parameter multimodal models claim to revolutionize on-device AI. The benchmarks look impressive, but real-world testing reveals a more nuanced story about the true cost of efficiency.

#lfm25#liquid-ai#multimodal-models...