Comment

Comment by formerly_proven

Open weight models are kind of weird because they're usually just a bunch of weights and maybe a chat template, not even test vectors or anything. Quite odd, and probably one of the reasons why vllm, sglang, the llamas etc. so often produce buggy output. It's actually quite strange how little attention is given to verifying these setups work.

Discussion

Live from the official Hacker News API

A modern reader for Hacker News · SvelteKit + Skeleton