News for hackers. Stuff that matters.

A comment on Ask HN: Has anyone replaced Claude/GPT with a local model for daily coding?. Parent comment

Re:Ask HN: Has anyone replaced Claude/GPT with a local model... by verdverm on Monday June 15, 2026 @11:43PM (#48550311)
There is a bug in llama-cpp for qwen/gemma models, use vLLM instead
[ Parent | on HN ]
Re:Ask HN: Has anyone replaced Claude/GPT with a local model... by pdyc on Tuesday June 16, 2026 @01:06AM (#48550845)
what bug and it affects what?
[ Parent | on HN ]
Re:Ask HN: Has anyone replaced Claude/GPT with a local model... by verdverm on Tuesday June 16, 2026 @11:19AM (#48556630)
it's a prompt cache invalidation bug that causes all input to be reprocessed instead of getting preloaded

There are other reasons to prefer vllm to llama-cpp as well

[ Parent | on HN ]