Running llama.cpp rather than vLLM, it's happy enough to run the FP8 variant wit... | Hacker News

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

		justaboutanyone 15 days ago \| parent \| context \| favorite \| on: Qwen3-Coder-Next Running llama.cpp rather than vLLM, it's happy enough to run the FP8 variant with 200k+ context using about 90GB vram

cmrdporcupine 15 days ago [–]

yeah, what did you get for tok/sec there though? Memory bandwidth is the limitation with these devices. With 4 bit I didn't get over 35-39 tok/sec, and averaged more like 30 when doing actual tool use with opencode. I can't imagine fp8 being faster.

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