Repository navigation

[None][feat] MiniMax-M3 NVFP4 low-latency decode megakernel for modeling_v2 for Blackwell - #20306

Open

peihu-nv wants to merge 32 commits into

NVIDIA:mainfrom

peihu-nv:peihengh/m3-megakernel

Open

[None][feat] MiniMax-M3 NVFP4 low-latency decode megakernel for modeling_v2 for Blackwell#20306
peihu-nv wants to merge 32 commits into
NVIDIA:mainfrom
peihu-nv:peihengh/m3-megakernel

Commits on Oct 15, 2026
Commits on Oct 19, 2026
Commits on Oct 20, 2026
Commits on Oct 21, 2026
- [None][chore] MiniMax-M3 modeling_v2: standard route interface, layout guard, target kernels under ops
  peihu-nv
  committed
  300c1c1
Commits on Oct 22, 2026
Commits on Oct 26, 2026
- [None][chore] Merge main 3205aeaedb into the MiniMax-M3 megakernel PR
  peihu-nv
  committed
  eea8a02
Commits on Oct 27, 2026
Commits on Oct 28, 2026
- [None][fix] Merge main and repair M3 CI test coverage
  peihu-nv
  committed
  ee47965
- [None][fix] Merge main preserving M3 and Kimi registrations
  peihu-nv
  committed
  6505d7f
- [None][fix] Isolate M3 fake registrations from stock imports
  peihu-nv
  committed
  7362f43
- [None][fix] Keep shared M3 fake ops in the kernel package
  peihu-nv
  committed
  53d48ee
Commits on Oct 29, 2026