#mobile inference11/10/2025
LFM2-8B-A1B: Liquid AI’s Sparse MoE That Runs a 1.5B Active Path on Phones
'Liquid AI released LFM2-8B-A1B, an 8.3B-parameter sparse Mixture-of-Experts designed for phones and edge devices that activates about 1.5B parameters per token to balance latency and capacity.'