Those are really nice numbers. With that t/s, no network latency or queueing it ... | Hacker News

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

		redox99 1 day ago \| parent \| context \| favorite \| on: Benchmarking Qwen3.8 27B quantizations: 4-bit hold... Those are really nice numbers. With that t/s, no network latency or queueing it must feel much snappier than cloud models.
		help

skolos 1 day ago [–]

it is, unless it set to xhigh - it really likes generating tons of tokens for its thinking. unfortunately, for decently reliable coding results you want it on xhigh ...

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