Show HN: Reame – a CPU inference server that gets faster as it runs
LLM inference serverCPU-firstllama.cppdisk KV cacheself-regulating speculationgeneration archive
Author: targetbridge
Date: 7/11/2026
Article Summary:
Reame is a CPU-first LLM inference server built on llama.cpp, designed to optimize performance on cheap CPU hardware by utilizing a disk KV cache, self-regulating speculation, and a generation archive.