[TRTLLM-8201][feat] Nemotron H MoE Sharding#8744

Merged

lucaslie merged 9 commits intoNVIDIA:mainfrom

nv-auto-deploy:ll/gk/sharding_mamba_rebased

Nov 5, 2025

Commits on Oct 30, 2025

rebased/cleaned up
lucaslie
committed
Fixed sharding for quantized linear models

greg-kwasniewski1
authored and
lucaslie
committed
Added missing i_any_lin_node

greg-kwasniewski1
authored and
lucaslie
committed
base_tp_plan patch for nemotron h moe
lucaslie
committed
fix unit test failures
lucaslie
committed
small config bug fix
lucaslie
committed
Fixed QKVO col-row sharding

greg-kwasniewski1
authored and
lucaslie
committed
Fixed quantized sharding

greg-kwasniewski1
authored and
lucaslie
committed

Commits on Nov 3, 2025

Merge branch 'main' into ll/gk/sharding_mamba_rebased
suyoggupta
authored