2 Commits

Author SHA1 Message Date
Nick Hill
8e7a891602
[BugFix] Fix spec decoding max_tokens scheduling perf issue (#29542)
Signed-off-by: Nick Hill <nhill@redhat.com>
2025-11-28 20:52:23 +08:00
usberkeley
4a8d6bd168
Fix cu_num_generated_tokens slicing logic in LogprobsLists.slice() method (#28214)
Signed-off-by: Bradley <bradley.b.pitt@gmail.com>
2025-11-09 19:11:46 +00:00