Feature: nemotron_omni decode is bound by 2944 conditional expert subgraphs per token (about 50 ms/token for a 3B-active model)

オープン
  • 何か問題がありますか？
#2,060 コメント 0 件 リアクション 0 件 担当者 0 名 GitHub で見る

メンテナーはふだん 1 日以内に返信

まだ誰も着手していません。

評価

難易度
5/5
見積もり時間
1週間以上
初心者へのやさしさ
35/100
issue の種類
機能追加
明瞭さ
おおむね明確
活発さ
活発
技術スタック
python, pytorch
領域
machine-learning

調査の方向性

Start with families/nemotron_omni/moe.py and inspect how each routed expert is built as an IIfConditional subgraph. Compare this with another family’s gathered MoE implementation, then investigate whether the selected experts’ weights can be run as a grouped or batched GEMM. Done means reducing per-token decode time and decoder plan and build size; the issue gives B200/B300 decode measurements for comparison.

索引モデルが issue の本文から書いたものです。

説明

enhancement

Request type: Improvement to existing behavior

Problem and use case

families/nemotron_omni builds each MoE layer of the Nemotron-H language model as 128 IIfConditional expert subgraphs (moe.py, one per routed expert; top-6 run). With 23 MoE layers that is 2944 conditional expert subgraphs per decode step, and the decoder plan has about 25k to 64k layers.

On the official NVFP4 checkpoint (the manifests' nvidia/Nemotron-3-Nano-Omni-30B-A3B-Reasoning-NVFP4), decode takes about 48 to 53 ms per token on B200 / B300 (about 20 tokens/s), for a model with about 3B active parameters (reading 3B parameters at about 8 TB/s would take well under 1 ms). Prefill is token by token through the same step, so it has the same cost per token.

The step time does not depend on the weight precision, which shows the step is bound by the conditional structure and kernel count, not by weight reads (decoder only, TensorRT 11.4, batch 1, ms per token at cache positions 100 / 2000):

GPU fp16 (from the BF16 checkpoint, built with the family decoder code) FP8 (official FP8 checkpoint) NVFP4 (as on main)
B200 47.7 / 47.8 48.3 / 48.4 49.7 / 49.8
B300 48.9 / 48.8 48.9 / 48.7 49.7 / 49.6

Decoder build time is also large: 24 to 52 minutes on B200 / B300 and about 101 minutes on H100 for NVFP4.

Proposed outcome

Build the routed experts of each MoE layer without one conditional subgraph per expert, for example a gathered/batched expert formulation (gather the selected experts' weights and run them as one grouped or batched GEMM, as other families do with a gathered MoE), so the decode step time scales with the active weights and the plan and build time shrink.

Model and target details

nvidia/Nemotron-3-Nano-Omni-30B-A3B-Reasoning-NVFP4 @ 16993199e436da4ba75ddc410855f87e0d996ee6 (the nemotron-omni-nvfp4-native manifest). Language model: 52 layers (23 Mamba2, 23 MoE with 128 routed experts top-6 plus 1 shared, 6 attention), hidden 2688. TensorRT 11.4; B200 / B300 (also seen on H100).

Alternatives considered

FP8 weights (official FP8 checkpoint) and fp16 weights give the same step time as NVFP4 (table above), so weight precision does not help while the MoE stays conditional.

Additional context

Found during an FP8 study of this family (not merged). Separate note for anyone building a Hugging Face reference for Nemotron-H: without mamba-ssm, the checkpoint's own PyTorch fallback in modeling_nemotron_h.py diverges after the first 128-token chunk (KL about 1e-8 on positions 0 to 127, 0.07 to 0.39 after); a plain sequential scan matches the fp16 engine (KL 0.0032).

主要言語
Python
スター
288
フォーク
73
平均マージ
1日 3時間
マージ済み PR（30日）
273

環境構築

はじめの一歩

  1. issue を最後まで読み、次にプロジェクトのコントリビューションガイドを読みます。
  2. 着手することを issue にコメントします — 二人が同じ作業をするのを防げます。
  3. リポジトリをフォークし、ブランチを切って変更します。
  4. issue 番号を参照したプルリクエストを送ります。

NVIDIA/TensorRT-Model-Connect のほかの issue

NVIDIA/TensorRT-Model-Connect の issue をすべて見る

似ている issue

Python の issue をもっと見る

新しい issue をメールで受け取る

初心者向けの GitHub issue を短くまとめたダイジェスト。