Added Nemotron-V3 support to benchmark.py and improved MFU on SQuAD#1308
Open
ethanhharrison wants to merge 3 commits intoNVIDIA-NeMo:mainfrom
Open
Added Nemotron-V3 support to benchmark.py and improved MFU on SQuAD#1308ethanhharrison wants to merge 3 commits intoNVIDIA-NeMo:mainfrom
ethanhharrison wants to merge 3 commits intoNVIDIA-NeMo:mainfrom
Conversation
Signed-off-by: Zhiyu Li <zhiyul@NVIDIA.com> perf lab Signed-off-by: Zhiyu Li <zhiyul@NVIDIA.com> perf lab (ep) Signed-off-by: Zhiyu Li <zhiyul@NVIDIA.com> better rope cache Signed-off-by: Zhiyu Li <zhiyul@NVIDIA.com> arithmetic intensity lab arithmetic intensity lab fix arithmetic intensity lab fix arithmetic intensity lab fix arithmetic intensity lab fix Revert "arithmetic intensity lab fix" This reverts commit b9a2c57. arithmetic intensity lab fix arithmetic intensity answer Signed-off-by: Zhiyu Li <zhiyul@NVIDIA.com> mixtral perf lab mixtral perf lab mixtral perf lab fix mixtral perf lab fix mixtral perf lab fix mixtral perf lab fix update grade_assignment update lab update Signed-off-by: Zhiyu Li <zhiyul@nvidia.com> arithmetic intensity answer Signed-off-by: Zhiyu Li <zhiyul@NVIDIA.com> remove reference remove reference Signed-off-by: Zhiyu Li <zhiyul@NVIDIA.com>
ethanhharrison
requested review from
HuiyingLi,
ZhiyuLi-Nvidia,
adil-a,
akoumpa and
hemildesai
as code owners
ethanhharrison
changed the title
ethanhharrison
changed the title
akoumpa
reviewed
Feb 17, 2026
| self.device_mesh = self._get_device_mesh() | ||
| # Always create the MoE mesh, even when ep_size == 1. | ||
| # This allows DeepEP to still initialize its EP process groups/dispatcher on a 1-way EP axis. | ||
| self.moe_mesh = self._get_moe_mesh() |
Contributor
There was a problem hiding this comment.
Thanks a lot for the PR @ethanhharrison !
We recently refactored the distributed part a bit, so I was thinking it would be better to rebase this PR on latest main before we proceed with the review/CI. Please let me know what you think.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
What does this PR do ?
Improves the MFU of Nemotron V3 Nano on the SQuAD finetuning task. Also allows for Nemotron V3 Nano support for the benchmark.py script.
Changelog
Removed the checks for sequence length and local batch size in benchmark.py which cause models other than Mixtral-8x7B to crash.
Increased the local batch size and EP for the Nemotron V3 Nano SQuAD finetuning config.
Added settings for the benchmark to Nemotron V3 Nano's config.