← Back to context Comment by jacquesm 8 hours ago How fast is it? 2 comments jacquesm Reply wincy 3 hours ago Parent already responded but just for reference an RTX 5090 with Ninfer hits 160 tokens/second with qwen 3.8 27B which is very usable. kristjansson 7 hours ago With MTP and FP4 I max out at 30ish t/s on mine. Without MTP or in regimes where the drafter performs poorly it’s about 10 t/s. FP8 is about half that
wincy 3 hours ago Parent already responded but just for reference an RTX 5090 with Ninfer hits 160 tokens/second with qwen 3.8 27B which is very usable.
kristjansson 7 hours ago With MTP and FP4 I max out at 30ish t/s on mine. Without MTP or in regimes where the drafter performs poorly it’s about 10 t/s. FP8 is about half that
Parent already responded but just for reference an RTX 5090 with Ninfer hits 160 tokens/second with qwen 3.8 27B which is very usable.
With MTP and FP4 I max out at 30ish t/s on mine. Without MTP or in regimes where the drafter performs poorly it’s about 10 t/s. FP8 is about half that