Issues / #1080

#1080 RTX 5000 PRO slow unsloth ud iq4_xs fix

open · @ukrolelo · 4 comments · View on GitHub

NVIDIA / CUDAModels & quants

Description

I didin't read what is this settings but i removed this
  "--resident-budget-gib",
  "55",

from the json and now it's flying, otherwise it was dead slow. 262K context, parallel 2, vision cpu
usage 47.1/47.8 VRAM

Related on strata.com

Editorial links to help you install, pick models, or read release notes — not part of the upstream thread.