NVLS all-reduce · 4 GPUs

A = ⊕aᵢ B = ⊕bᵢ C = ⊕cᵢ D = ⊕dᵢ = selected NCCL reduction
UC = per-GPU backing replica MC = one multicast address over 4 replicas 3rd-gen NVSwitch · NVLink 4+
Input Step 0 of 4 · One tensor per GPU
One logical chunk group; NCCL pipelines and overlaps these phases across chunks and channels. Registered buffers can bypass staging copies.