picchio

picchio

Catches silent CPU fallback and mislabeled tok/s in local LLMs. llama.cpp and ollama, one file, no deps.

18 319 HN points
MIT
last commit 2026-07-15
Source
Share:

About

Catch local LLMs spilling into the CPU. One run shows GPU placement, prefill, decode, memory, power and a verdict.

picchio website preview

Languages

Contributors1

No features listed.

Comments Theme
slug: picchio