+1 using llama.cpp Vulkan releases with the Qwen models - runs much better than ... | Hacker News

Hacker Timesnew | past | comments | ask | show | jobs | submit

ndom91 60 days ago | parent | context | favorite | on: Ask HN: Has anyone replaced Claude/GPT with a loca...

+1 using llama.cpp Vulkan releases with the Qwen models - runs much better than the ROCm releases.

I'll have to give the preserve_thinking a shot.

jderekw 59 days ago [–]

Thanks for sharing have been running ROCm primarily with Qwen 3.6 and Qwen Coder, on the runs much better statement is that a stability, performance or other capability your experiencing?

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact