Skip to content

feat: Enable vLLM cudagraphs #498

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom
Merged
merged 4 commits into from
Jul 3, 2025
Merged

feat: Enable vLLM cudagraphs #498

merged 4 commits into from
Jul 3, 2025

Conversation

jiemingz
Copy link
Contributor

@jiemingz jiemingz commented Jun 10, 2025 •
edited
Loading

Addresses: !186

image image image

The generation throughput shows about ~3% speedup for llama8b on 4 nodes

What does this PR do ?

Add a one line overview of what this PR aims to accomplish.

Issues

List issues that this PR closes (syntax):

Usage

  • You can potentially add a usage example below
# Add a code snippet demonstrating how to use this

Before your PR is "Ready for review"

Pre checks:

  • Make sure you read and followed Contributor guidelines
  • Did you write any new necessary tests?
  • Did you run the unit tests and functional tests locally? Visit our Testing Guide for how to run tests
  • Did you add or update any necessary documentation? Visit our Document Development Guide for how to write, build and test the docs.

Additional Information

  • ...

@jiemingz jiemingz changed the title Draft: Enable vLLM cudagraphs Enable vLLM cudagraphs Jun 11, 2025
@jiemingz jiemingz self-assigned this Jun 11, 2025
parthchadha
parthchadha reviewed Jun 11, 2025
View reviewed changes
@jiemingz jiemingz changed the title Enable vLLM cudagraphs feat: Enable vLLM cudagraphs Jun 13, 2025
parthchadha
parthchadha requested changes Jun 17, 2025
View reviewed changes
@parthchadha
Copy link
Contributor

@jiemingz can you also add timing plot to the MR description showing benefits of enabling cuda graphs vs not.

parthchadha
parthchadha previously approved these changes Jun 17, 2025
View reviewed changes
SahilJain314
SahilJain314 previously approved these changes Jun 26, 2025
View reviewed changes
parthchadha
parthchadha previously approved these changes Jun 26, 2025
View reviewed changes
@parthchadha parthchadha added this pull request to the merge queue Jun 26, 2025
@github-merge-queue github-merge-queue bot removed this pull request from the merge queue due to failed status checks Jun 27, 2025
@SahilJain314
Copy link
Contributor

Unit test failure here with the eager key missing: @jiemingz
E File "/opt/nemo-rl/nemo_rl/models/generation/vllm.py", line 336, in init
E enforce_eager=self.cfg["vllm_cfg"]["enforce_eager"],
E ~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^
E KeyError: 'enforce_eager'

@jiemingz jiemingz dismissed stale reviews from parthchadha and SahilJain314 via 3c08662 June 30, 2025 16:02
@jiemingz jiemingz force-pushed the jiemingz/vllm_cg branch 2 times, most recently from 3c08662 to 7bb9f3d Compare June 30, 2025 16:08
parthchadha
parthchadha previously approved these changes Jul 1, 2025
View reviewed changes
@parthchadha parthchadha added this pull request to the merge queue Jul 1, 2025
@github-merge-queue github-merge-queue bot removed this pull request from the merge queue due to failed status checks Jul 1, 2025
parthchadha
parthchadha previously approved these changes Jul 2, 2025
View reviewed changes
parthchadha
parthchadha previously approved these changes Jul 2, 2025
View reviewed changes
@parthchadha parthchadha added this pull request to the merge queue Jul 2, 2025
@github-merge-queue github-merge-queue bot removed this pull request from the merge queue due to a conflict with the base branch Jul 2, 2025
parthchadha
parthchadha previously approved these changes Jul 2, 2025
View reviewed changes
@parthchadha parthchadha added this pull request to the merge queue Jul 2, 2025
github-merge-queue bot pushed a commit that referenced this pull request Jul 2, 2025
@jiemingz
feat: Enable vLLM cudagraphs (#498)
b1ab244 
Signed-off-by: Jimmy Zhang <jiemingz@nvidia.com>
Signed-off-by: Jimmy Zhang <133159885+jiemingz@users.noreply.github.com>
@github-merge-queue github-merge-queue bot removed this pull request from the merge queue due to failed status checks Jul 2, 2025
parthchadha
parthchadha previously approved these changes Jul 2, 2025
View reviewed changes
@parthchadha parthchadha added this pull request to the merge queue Jul 2, 2025
@github-merge-queue github-merge-queue bot removed this pull request from the merge queue due to a conflict with the base branch Jul 3, 2025
jiemingz and others added 3 commits July 3, 2025 06:25
@jiemingz
vllm CG
d9e3f40 
Signed-off-by: Jimmy Zhang <jiemingz@nvidia.com>
@jiemingz
Update grpo-gemma3-27b-it-16n8g-fsdp2tp8sp-actckpt-long.yaml
1ac69d2 
Signed-off-by: Jimmy Zhang <133159885+jiemingz@users.noreply.github.com>
@jiemingz
add key
e545a48 
Signed-off-by: Jimmy Zhang <jiemingz@nvidia.com>
@jiemingz jiemingz force-pushed the jiemingz/vllm_cg branch from 2610899 to e545a48 Compare July 3, 2025 13:27
@parthchadha parthchadha added this pull request to the merge queue Jul 3, 2025
@github-merge-queue github-merge-queue bot removed this pull request from the merge queue due to failed status checks Jul 3, 2025
@parthchadha parthchadha enabled auto-merge July 3, 2025 19:14
@parthchadha parthchadha added this pull request to the merge queue Jul 3, 2025
Merged via the queue into main with commit 6ca1588 Jul 3, 2025
13 of 14 checks passed
@parthchadha parthchadha deleted the jiemingz/vllm_cg branch July 3, 2025 21:22
therealnaveenkamal pushed a commit to therealnaveenkamal/RL that referenced this pull request Jul 7, 2025
@jiemingz @parthchadha @therealnaveenkamal
feat: Enable vLLM cudagraphs (NVIDIA-NeMo#498)
4eb4bf5 
Signed-off-by: Jimmy Zhang <jiemingz@nvidia.com>
Signed-off-by: Jimmy Zhang <133159885+jiemingz@users.noreply.github.com>
Co-authored-by: Parth Chadha <pchadha@nvidia.com>
jialei777 pushed a commit to jialei777/nemo-rl that referenced this pull request Jul 23, 2025
@jiemingz @parthchadha @jialei777
feat: Enable vLLM cudagraphs (NVIDIA-NeMo#498)
ab84468 
Signed-off-by: Jimmy Zhang <jiemingz@nvidia.com>
Signed-off-by: Jimmy Zhang <133159885+jiemingz@users.noreply.github.com>
Co-authored-by: Parth Chadha <pchadha@nvidia.com>
Signed-off-by: Jialei Chen <jialeic@google.com>
KiddoZhu pushed a commit that referenced this pull request Jul 28, 2025
@jiemingz @parthchadha @KiddoZhu
feat: Enable vLLM cudagraphs (#498)
8a054a4 
Signed-off-by: Jimmy Zhang <jiemingz@nvidia.com>
Signed-off-by: Jimmy Zhang <133159885+jiemingz@users.noreply.github.com>
Co-authored-by: Parth Chadha <pchadha@nvidia.com>
FannYYW pushed a commit to xxman-google/NeMo-RL that referenced this pull request Aug 5, 2025
@jiemingz @parthchadha
feat: Enable vLLM cudagraphs (NVIDIA-NeMo#498)
990a569 
Signed-off-by: Jimmy Zhang <jiemingz@nvidia.com>
Signed-off-by: Jimmy Zhang <133159885+jiemingz@users.noreply.github.com>
Co-authored-by: Parth Chadha <pchadha@nvidia.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@parthchadha parthchadha parthchadha approved these changes

@SahilJain314 SahilJain314 SahilJain314 left review comments

@terrykong terrykong Awaiting requested review from terrykong

Assignees

@jiemingz jiemingz

Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
3 participants
@jiemingz @parthchadha @SahilJain314