forked from ggml-org/llama.cpp
-
-
Notifications
You must be signed in to change notification settings - Fork 41
Pull requests: Anbeeld/beellama.cpp
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
fix(cuda): KVarN decode combine kernel crashes at ≥768K context (missing shared-mem opt-in)
#116
opened Aug 2, 2026 by
chimpera
Loading…
reasoning-budget: staged injection (intro/soft/hard) + optional soft grace period
#106
opened Jul 25, 2026 by
Andgihat
Loading…
ProTip!
Add no:assignee to see everything that’s not assigned.