Skip to content

[PyTorch] Quantize distributed BF16 weights in the fused grouped MLP - #3695

Closed
xrennvidia wants to merge 2 commits into
NVIDIA:mainfrom
xrennvidia:xren/fused_grouped_mlp_gtp_bf16_weight_requant
Closed

[PyTorch] Quantize distributed BF16 weights in the fused grouped MLP#3695
xrennvidia wants to merge 2 commits into
NVIDIA:mainfrom
xrennvidia:xren/fused_grouped_mlp_gtp_bf16_weight_requant

Commits

  1. Commits on Oct 18, 2026