Agentic Kernel Optimization at the scale of the web
LFM2.5 230M from Liquid AI running entirely locally in the browser on WebGPU, with every kernel optimized by Fable 5 and Opus 4.8. Private by design: no data ever leaves your machine.
LFM2.5 230M from Liquid AI running entirely locally in the browser on WebGPU, with every kernel optimized by Fable 5 and Opus 4.8. Private by design: no data ever leaves your machine.
LFM2.5 runs entirely on your device.
Kernels are the low-level GPU programs that do the model's actual math — the matrix multiplications, attention, and normalization behind every token. And how well they're optimized can dramatically speed up inference.