Skip to content

Conversation

dmitry-tokarev-nv
Copy link
Contributor

@dmitry-tokarev-nv dmitry-tokarev-nv commented Aug 19, 2025

Overview:

Upgrade vllm in pyproject.toml and flashinfer version and update comments/references
This is a follow up on #2509

Details:

Where should the reviewer start?

Related Issues: (use one of the action keywords Closes / Fixes / Resolves / Relates to)

  • closes GitHub issue: #xxx

Summary by CodeRabbit

  • New Features

    • None.
  • Chores

    • Upgraded vLLM to 0.10.1 across build and optional dependencies.
    • Updated FlashInf to v0.2.11 in build and install defaults.
  • Documentation

    • Install script help now reflects current default values dynamically.
    • Updated inline reference link in configuration comments.
  • Refactor

    • None.
  • Bug Fixes

    • None.

Copy link

copy-pr-bot bot commented Aug 19, 2025

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.

Copy link
Contributor

coderabbitai bot commented Aug 19, 2025

Walkthrough

Bumps vLLM to 0.10.1 across config and packaging, updates FlashInf reference, adjusts install script defaults/help to reference variables, and updates a documentation URL in a Python comment. No functional code or control flow changes.

Changes

Cohort / File(s) Summary
vLLM version bump
container/Dockerfile.vllm, pyproject.toml
Update vLLM from 0.10.0 to 0.10.1 in Docker build and optional dependency specification.
FlashInf version bump
container/Dockerfile.vllm, container/deps/vllm/install_vllm.sh
Update FLASHINF_REF from v0.2.8rc1 to v0.2.11 in Dockerfile and install script default.
Install script help defaults via variables
container/deps/vllm/install_vllm.sh
Help/usage text now renders defaults from variables (VLLM_REF, MAX_JOBS, INSTALLATION_DIR, DEEPGEMM_REF, FLASHINF_REF, TORCH_BACKEND); no logic changes.
Doc comment URL update
components/backends/vllm/src/dynamo/vllm/args.py
Update reference URL in comment for NIXL side-channel port calculation (v0.10.0 → v0.10.1); no code changes.

Estimated code review effort

🎯 2 (Simple) | ⏱️ ~10 minutes

Possibly related PRs

Poem

I twitch my nose at tags that climb,
From point-oh to point-oh-one in rhyme.
Flashing fast, the refs align,
Scripts now echo defaults fine.
A hop, a bump, no logic bent—
Carrots cached, I’m dependency-sent! 🥕✨

Tip

🔌 Remote MCP (Model Context Protocol) integration is now available!

Pro plan users can now connect to remote MCP servers from the Integrations page. Connect with popular remote MCPs such as Notion and Linear to add more context to your reviews and chats.


Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share
🪧 Tips

Chat

There are 3 ways to chat with CodeRabbit:

  • Review comments: Directly reply to a review comment made by CodeRabbit. Example:
    • I pushed a fix in commit <commit_id>, please review it.
    • Open a follow-up GitHub issue for this discussion.
  • Files and specific lines of code (under the "Files changed" tab): Tag @coderabbitai in a new review comment at the desired location with your query.
  • PR comments: Tag @coderabbitai in a new PR comment to ask questions about the PR branch. For the best results, please provide a very specific query, as very limited context is provided in this mode. Examples:
    • @coderabbitai gather interesting stats about this repository and render them as a table. Additionally, render a pie chart showing the language distribution in the codebase.
    • @coderabbitai read the files in the src/scheduler package and generate a class diagram using mermaid and a README in the markdown format.

Support

Need help? Create a ticket on our support page for assistance with any issues or questions.

CodeRabbit Commands (Invoked using PR/Issue comments)

Type @coderabbitai help to get the list of available commands.

Other keywords and placeholders

  • Add @coderabbitai ignore anywhere in the PR description to prevent this PR from being reviewed.
  • Add @coderabbitai summary to generate the high-level summary at a specific location in the PR description.
  • Add @coderabbitai anywhere in the PR title to generate the title automatically.

Status, Documentation and Community

  • Visit our Status Page to check the current availability of CodeRabbit.
  • Visit our Documentation for detailed information on how to use CodeRabbit.
  • Join our Discord Community to get help, request features, and share feedback.
  • Follow us on X/Twitter for updates and announcements.

Copy link
Contributor

@coderabbitai coderabbitai bot left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Actionable comments posted: 2

🧹 Nitpick comments (1)
components/backends/vllm/src/dynamo/vllm/args.py (1)

212-216: Assertion message references kv_port but checks side_channel_port

The assert checks side_channel_port is set, but the error message says “Must set the kv_port...”. This can mislead operators debugging misconfiguration.

Apply this diff to correct the message:

-    assert (
-        config.side_channel_port is not None
-    ), "Must set the kv_port, use configure_ports_with_etcd"
+    assert (
+        config.side_channel_port is not None
+    ), "Must set side_channel_port, use configure_ports_with_etcd"
📜 Review details

Configuration used: .coderabbit.yaml
Review profile: CHILL
Plan: Pro

💡 Knowledge Base configuration:

  • MCP integration is disabled by default for public repositories
  • Jira integration is disabled by default for public repositories
  • Linear integration is disabled by default for public repositories

You can enable these sources in your CodeRabbit configuration.

📥 Commits

Reviewing files that changed from the base of the PR and between cae5822 and 7118d27.

📒 Files selected for processing (4)
  • components/backends/vllm/src/dynamo/vllm/args.py (1 hunks)
  • container/Dockerfile.vllm (1 hunks)
  • container/deps/vllm/install_vllm.sh (2 hunks)
  • pyproject.toml (1 hunks)
🔇 Additional comments (3)
pyproject.toml (1)

56-60: Pin update to vLLM 0.10.1 looks good

The optional dependency bump to vllm==0.10.1 aligns with the Dockerfile notes and install script changes. NIXL constraint remains <=0.4.1, consistent with the Dockerfile’s NIXL_REF=0.4.1.

components/backends/vllm/src/dynamo/vllm/args.py (1)

172-176: Doc reference updated to v0.10.1

The URL now points to the corresponding line in vLLM 0.10.1. No code behavior change.

container/deps/vllm/install_vllm.sh (1)

29-31: FlashInfer default bumped to v0.2.11

Default aligns with the Dockerfile ARG and PR objective. Looks good.

@dmitry-tokarev-nv
Copy link
Contributor Author

/ok to test 7118d27

@pull-request-size pull-request-size bot added size/M and removed size/S labels Aug 19, 2025
Copy link
Contributor

@alec-flowers alec-flowers left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Examples are working for me, just approved MR. Two caveats:

  • lm_cache example doesn't work but this is known as we don't install it into the arm64 container

  • I'm not currently testing dsr1 as we haven't verified functionality in previous releases for arm64

@dmitry-tokarev-nv
Copy link
Contributor Author

/ok to test 10148a7

@alec-flowers alec-flowers enabled auto-merge (squash) August 19, 2025 19:45
@alec-flowers alec-flowers merged commit 5f57ea5 into main Aug 19, 2025
12 of 13 checks passed
@alec-flowers alec-flowers deleted the dtokarev-finish-upgrade-vllm-0.10.1 branch August 19, 2025 19:48
@coderabbitai coderabbitai bot mentioned this pull request Aug 22, 2025
hhzhang16 pushed a commit that referenced this pull request Aug 27, 2025
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants