perf(cohere2): fuse the parallel-block residual adds with the next norm - #1948

Merged

inureyes merged 3 commits into

mainfrom

perf/cohere2-fused-residual-norm

Sep 23, 2026

Commits on Sep 21, 2026

perf(cohere2): fuse the parallel-block residual adds into one kernel
inureyes
committed
c5e3a13
perf(cohere2): fuse the residual add with the next LayerNorm
inureyes
committed
d392ba8

Commits on Sep 23, 2026

Merge remote-tracking branch 'origin/main' into perf/cohere2-fused-residual-norm
inureyes
committed
edec8cd