View source for Qwen3.8 27B on 1x 8 GB (W4A16 vLLM)/Confirmation 1

{{Confirmation
|page=Qwen3.8 27B on 1x 8 GB (W4A16 vLLM)
|confirmed_by=chonsworth
|date=2026-10-05
|cards=8 GB
|result=Partly reproduced
|numbers=About 3.7k tok/s prefill at about 100 tok/s decode, against the published 1.9k / 100. Never run to 1M context.
|notes=Replying to kellertreppe123's post of the recipe, on the same model on one card. He did not say whether he used the recipe as written; his prefill is higher than published and his decode matches. {{Source|handle=chonsworth|date=2026-10-05|link=https://discord.com/channels/1518634379604136178/1528826330454429706/1556775994692731011}}
}}

{{ImportFooter}}