Local LLM: LFM2.5 on a Laptop CPU (PyTorch)

One generation from LiquidAI’s LFM2.5-2.6B, run on a laptop CPU in bfloat16 with PyTorch: 1,487 tokens at 7.0 tokens per second. This is the slow starting point for the next two.

Last updated 2026-08-05 00:03 UTC · made on claude.ai

Model, effort and cost weren't recorded for this one: it was made on claude.ai, which doesn't keep them with the page.