-
-
Notifications
You must be signed in to change notification settings - Fork 140
Pull requests: turboderp-org/exllamav3
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
Validate reconstruct kernel indices before dispatch
#310
opened Aug 31, 2026 by
professor-moody
Loading…
Speed up Qwen MTP decoding by 22% with a selected vocabulary head
#303
opened Aug 22, 2026 by
keldenl
Loading…
PoC: per-layer split or fused-uniform QKV topology
#299
opened Aug 20, 2026 by
malaiwah
Contributor
Loading…
exl3_mgemm: fix out-of-bounds slot list, and allow multi-token reduction with expert-range filtering
#290
opened Aug 17, 2026 by
arbi-dev
Loading…
Allow using multiple draft models at once
#261
opened Jul 28, 2026 by
Ednaordinary
•
Draft
2 of 5 tasks
[kernel] Persistent route-packed (route_block x n_tile) schedule for exl3 fused MoE (sm120)
#246
opened Jul 15, 2026 by
brandonmmusic-max
•
Draft
ProTip!
What’s not been updated in a month: updated:<2026-07-31.