Skip to content

Fix CVE pkgs - #2440

Merged
jperez999 merged 7 commits into
NVIDIA:mainfrom
jperez999:fix-cve-pkgs
Aug 7, 2026
Merged

Fix CVE pkgs#2440
jperez999 merged 7 commits into
NVIDIA:mainfrom
jperez999:fix-cve-pkgs

Conversation

@jperez999

Copy link
Copy Markdown
Collaborator

Description

Checklist

  • I am familiar with the Contributing Guidelines.
  • New or existing tests cover these changes.
  • The documentation is up to date with these changes.

@jperez999
jperez999 requested review from a team as code owners August 7, 2026 16:07
@jperez999
jperez999 requested a review from ChrisJar August 7, 2026 16:07
@copy-pr-bot

copy-pr-bot Bot commented Aug 7, 2026

Copy link
Copy Markdown

This pull request requires additional validation before any workflows can run on NVIDIA's runners.

Pull request vetters can view their responsibilities here.

Contributors can view more details about this message here.

@greptile-apps

greptile-apps Bot commented Aug 7, 2026

Copy link
Copy Markdown
Contributor

Greptile Summary

The PR updates vulnerable dependencies and aligns the Linux local-inference stack around exact vLLM and FlashInfer versions.

  • Pins vLLM 0.25.1 with FlashInfer 0.6.13 in the published local extra.
  • Raises minimum versions for Ray, Transformers, NLTK, Pillow, and LiteLLM.
  • Updates the packaging assertion for the new LiteLLM minimum.
  • Removes Intel macOS jobs from the CI and integration-test matrices.

Confidence Score: 5/5

The PR appears safe to merge.

No blocking failure remains; the current published local extra exactly pins vLLM 0.25.1 and both FlashInfer packages to 0.6.13, resolving the previously reported open-ended GPU dependency stack.

Important Files Changed

Filename Overview
nemo_retriever/pyproject.toml Updates dependency constraints and completes the previously requested exact vLLM pin alongside the matching FlashInfer versions.
nemo_retriever/tests/test_service_packaging.py Updates the packaging test to assert the revised LiteLLM minimum.
.github/workflows/ci-pull-request.yml Removes the Intel macOS runner from the pull-request test matrix.
.github/workflows/integration-test-library-mode.yml Removes the Intel macOS runner from the library-mode integration matrix.

Reviews (5): Last reviewed commit: "fix formatting" | Re-trigger Greptile

Comment thread nemo_retriever/pyproject.toml Outdated
Comment thread nemo_retriever/pyproject.toml Outdated
@kheiss-uwzoo

kheiss-uwzoo commented Aug 7, 2026

Copy link
Copy Markdown
Collaborator

Docs follow-up for the macOS Intel drop (Randy / Slack):

@kheiss-uwzoo

Copy link
Copy Markdown
Collaborator

Install-guide draft is up: #2441 (pairs with changelog in draft #2434).

@jperez999
jperez999 merged commit 0513039 into NVIDIA:main Aug 7, 2026
8 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants