Skip to content

Generalized Tensor Parallelism (GTP) #3005

Open
fanshiqing wants to merge 5 commits into
NVIDIA:mainfrom
fanshiqing:gtp_release
Open

Generalized Tensor Parallelism (GTP) #3005
fanshiqing wants to merge 5 commits into
NVIDIA:mainfrom
fanshiqing:gtp_release

[fix] Respect per-op activation-offload markers in fused grouped MLP

4a12eb4
Select commit
Loading
Failed to load commit list.