Repository navigation

cuda: serve --spill-gib and --checkpoint-slots for Nemotron, the Qwen hybrids and Flash Next (#840, split from #1359) - #1413

Merged

ashhart merged 3 commits into

ashhart:mainfrom

jschmied:qwen-cuda-spill

Oct 28, 2026

Merged

cuda: serve --spill-gib and --checkpoint-slots for Nemotron, the Qwen hybrids and Flash Next (#840, split from #1359)#1413
ashhart merged 3 commits into
ashhart:mainfrom
jschmied:qwen-cuda-spill

Commits on Oct 28, 2026
- cuda: serve --spill-gib and --checkpoint-slots for Nemotron, the Qwen hybrids and Flash Next
  
  authored and
  ashhart
  committed
  374b39b
- cuda: a host test that CUDA's disk open hands --spill-gib's mode and --checkpoint-slots to the store
  
  jschmied
  authored and
  ashhart
  committed
  9f139b1
- Maintainer fix-up for #1413: the CUDA spill paragraph points up to the read-ahead paragraph
  
  ashhart
  and
  jschmied
  committed
  28ddca5