Agent skill · NVIDIA
nemo-mbridge-perf-moe-hardware-configs
Representative, point-in-time MoE training playbooks by hardware and model family. Use them as candidate seeds, then revalidate the exact runtime, semantics, topology, and steady-state throughput.
What it needs
About 5k tokens when loaded.
What this skill does
MoE Hardware Configuration Reference Stable docs: @docs/training/moe-optimization.md Card: @skills/nemo-mbridge-perf-moe-hardware-configs/card.yaml Quick Platform Playbook These rows are search seeds, not hardware defaults or throughput promises. Platform Candidates to screen after alltoall bring-up What usually matters most --- --- --- H100 DeepEP or HybridEP, explicit overlap, supported FP8 modes communication overlap, dispatcher/runtime compatibility, and PP efficiency B200 DeepEP or HybridEP, supported FP8 modes, careful PP layout container quality and tuned communication settings GB200 HybridEP, then profile-driven graphs and CPU cleanup host overhead, topology-aware dispatch, memory headroom GB300 HybridEP and the target container's lower-precision/kernel stack the same system interactions as GB200, with remeasurement required First Answer Checklist For hardware playbook questions, answer from these canonical rows before adding throughput caveats: Workload Hardware Dispatcher Layout --- --- --- --- DSV3 H100 DeepEP TP=2, EP=64, PP=8, VPP=4 DSV3 GB200/GB300 HybridEP TP=1, EP=64, PP=4, VPP=4 Qwen3 235B H100 alltoall + overlap in the current canonical recipe TP=2, EP=32, PP=8, VPP=4 Qwen3 235B GB200 HybridEP TP=1 or 2, EP=32-64, PP=4, VPP=unspecified Qwen3 30B 16×H100 HybridEP TP=1, EP=16, PP=1, plain EP overlap For Qwen3 235B on GB200, explicitly say VPP=unspecified; do not invent or extrapolate VPP=12 unless a measured row provides it. Treat TE-scoped CUDA graph scopes (attn, moerouter, moepreprocess) as profile-driven candidates, CUDADEVICEMAXCONNECTIONS selection, PYTORCHCUDAALLOCCONF=expandablesegments:True, NCCLGRAPHREGISTER=0, GB200/GB300 CPU-side tuning, and the warning not to cargo-cult tracker rows. Rounded Performance Bands These are intentionally rounded so the document stays durable as the tracker moves. Treat them as planning ranges, not exact promises. …
How to use it
Reference it in AdaL, Claude Code, Cursor or any coding agent — nothing to install:
@skills NVIDIA/nemo-mbridge-perf-moe-hardware-configs