News for hackers. Stuff that matters.
Re:Why do output tokens cost 5x more than input tokens? by ani17 on Monday April 20, 2026 @09:41AM (#47834196)
Author here. I wanted to understand what vLLM and llama.cpp are actually doing under the hood, but the codebases are massive. So I wrote a stripped down version from scratch to see the core ideas without the production complexity.

Code: https://github.com/Anirudh171202/WhiteLotus

Re:Why do output tokens cost 5x more than input tokens? by lazyMonkey69 on Monday April 20, 2026 @09:22AM (#47833953)
I think the paged attention part is a bit oversimplified. Nice read otherwise!