Introduction So you’ve built a €9,000 Grace–Hopper “desktop” (see: my previous post involving 16-million-degree GPU temperatures). Running llama.cpp benchmarks is fine, but the real test of local ...

Introduction Running large language models locally has always been a game of compromise. You either spend \$10,000+ on consumer GPUs that can barely handle 70B parameter models, or you dream about...

Speculative Article - caveat emptor No one knows how big AGI needs to be. The current consensus among the scaling-pilled crowd is “trillions of parameters and a nuclear power plant.” Maybe they’re...

Speculative Article - caveat emptor I was debugging a speech recognition algorithm last week when I had a weird thought: we’re about to make most of humanity economically redundant, and our respon...