EdgeBench

EdgeBench is an AI benchmarking tool that runs 134 tasks for at least 12 hours each to measure model performance at the edge. It is developed by Liquid AI to evaluate small models like LFM2.5-230M, which outperforms models four times its size.

Two-hop subgraph: this entity, every entity it directly relates to, and every entity those neighbors relate to. Drag a node, scroll to zoom, click to inspect — or click any neighbor and re-center the atlas there.