feat(router): add initial support for anthropic messages endpoint #775

nejch · 2025-12-25T23:24:46Z

This is a follow-up to vllm-project/vllm#21313, a simple initial PR adding support here to route requests to /v1/messages in vLLM itself without transforming requests here.

Hopefully we can later extend this to also route to Anthropic endpoints themselves, e.g. via bedrock, but that needs some healthcheck refactoring first.

Related to Anthropic tasks in #640
Also related to #278

BEFORE SUBMITTING, PLEASE READ THE CHECKLIST BELOW AND FILL IN THE DESCRIPTION ABOVE

Make sure the code changes pass the pre-commit checks.
Sign-off your commit by using -s when doing git commit
Try to classify PRs for easy understanding of the type of changes, such as [Bugfix], [Feat], and [CI].

Detailed Checklist (Click to Expand)

Thank you for your contribution to production-stack! Before submitting the pull request, please ensure the PR meets the following criteria. This helps us maintain the code quality and improve the efficiency of the review process.

PR Title and Classification

Please try to classify PRs for easy understanding of the type of changes. The PR title is prefixed appropriately to indicate the type of change. Please use one of the following:

[Bugfix] for bug fixes.
[CI/Build] for build or continuous integration improvements.
[Doc] for documentation fixes and improvements.
[Feat] for new features in the cluster (e.g., autoscaling, disaggregated prefill, etc.).
[Router] for changes to the vllm_router (e.g., routing algorithm, router observability, etc.).
[Misc] for PRs that do not fit the above categories. Please use this sparingly.

Note: If the PR spans more than one category, please include all relevant prefixes.

Code Quality

The PR need to meet the following code quality standards:

Pass all linter checks. Please use pre-commit to format your code. See README.md for installation.
The code need to be well-documented to ensure future contributors can easily understand the code.
Please include sufficient tests to ensure the change is stay correct and robust. This includes both unit tests and integration tests.

DCO and Signed-off-by

When contributing changes to this project, you must agree to the DCO. Commits must include a Signed-off-by: header which certifies agreement with the terms of the DCO.

Using -s with git commit will automatically add this header.

What to Expect for the Reviews

We aim to address all PRs in a timely manner. If no one reviews your PR within 5 days, please @-mention one of YuhanLiu11
, Shaoting-Feng or ApostaC.

Signed-off-by: Nejc Habjan <nejc.habjan@siemens.com>

gemini-code-assist · 2025-12-25T23:25:00Z

Summary of Changes

Hello @nejch, I'm Gemini Code Assist¹! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed!

This pull request introduces foundational support for handling Anthropic-style /v1/messages API requests within the vLLM router. It establishes the necessary routing infrastructure and utility functions to recognize and process these requests, marking a significant first step towards broader compatibility with Anthropic's API specifications. The current implementation focuses on routing requests without transformation, with future plans to extend functionality for direct Anthropic endpoint interaction.

Highlights

Anthropic Messages Endpoint Support: Initial support has been added for routing Anthropic-compatible /v1/messages requests directly within the vLLM router, laying the groundwork for future integrations.
New Anthropic Router: A dedicated anthropic_router.py file has been introduced, defining an APIRouter and a POST endpoint for /v1/messages to handle these specific requests.
Utility Updates: The ModelType enum in utils.py has been extended to include messages, along with corresponding URL and test payload generation for this new model type.

🧠 New Feature in Public Preview: You can now enable Memory to help Gemini Code Assist learn from your team's feedback. This makes future code reviews more consistent and personalized to your project's style. Click here to enable Memory in your admin console.

Using Gemini Code Assist

The full guide for Gemini Code Assist can be found on our documentation page, here are some quick tips.

Invoking Gemini

You can request assistance from Gemini at any point by creating a comment using either /gemini <command> or @gemini-code-assist <command>. Below is a summary of the supported commands on the current page.

Feature	Command	Description
Code Review	`/gemini review`	Performs a code review for the current pull request in its current state.
Pull Request Summary	`/gemini summary`	Provides a summary of the current pull request in its current state.
Comment	@gemini-code-assist	Responds in comments when explicitly tagged, both in pull request comments and review comments.
Help	`/gemini help`	Displays a list of available commands.

Customization

To customize Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a .gemini/ folder in the base of the repository. Detailed instructions can be found here.

Limitations & Feedback

Gemini Code Assist may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counter productive. You can react with 👍 and 👎 on @gemini-code-assist comments. If you're interested in giving your feedback about your experience with Gemini Code Assist for Github and other Google products, sign up here.

You can also get AI-powered code generation, chat, as well as code reviews directly in the IDE at no cost with the Gemini Code Assist IDE Extension.

Review the Privacy Notices, Generative AI Prohibited Use Policy, Terms of Service, and learn how to configure Gemini Code Assist in GitHub here. Gemini can make mistakes, so double check it and use code with caution. ↩

gemini-code-assist

Code Review

This pull request introduces initial support for the Anthropic messages API endpoint (/v1/messages) by adding a new router to forward these requests to the vLLM backend. The changes are well-structured and align with the existing codebase patterns. My review includes one suggestion to refactor a piece of duplicated code in the test payload generation logic to enhance maintainability. Overall, the implementation is solid and effectively accomplishes its goal.

gemini-code-assist · 2025-12-25T23:25:59Z

src/vllm_router/utils.py

+            case ModelType.messages:
+                return {
+                    "messages": [
+                        {
+                            "role": "user",
+                            "content": "Hello",
+                        }
+                    ],
+                    "temperature": 0.0,
+                    "max_tokens": 3,
+                }


The test payload for messages is identical to the one for chat. To avoid code duplication and improve maintainability, you can simplify this by reusing the payload logic from the chat model type.

case ModelType.messages: return ModelType.get_test_payload("chat")

Not sure this makes sense as we're actually inside get_test_payload here, so it wouldn't be that intuitive for humans. I can factor out the payload itself if needed to avoid duplication.

nejch · 2025-12-25T23:25:27Z

src/vllm_router/routers/anthropic_router.py

I'm adding a separate module just in case, so we can add future Anthropic-compatible endpoints here if needed.

nejch · 2025-12-25T23:30:25Z

src/vllm_router/utils.py

+            case ModelType.messages:
+                return {
+                    "messages": [
+                        {
+                            "role": "user",
+                            "content": "Hello",
+                        }
+                    ],
+                    "temperature": 0.0,
+                    "max_tokens": 3,
+                }


Not sure this makes sense as we're actually inside get_test_payload here, so it wouldn't be that intuitive for humans. I can factor out the payload itself if needed to avoid duplication.

nejch · 2025-12-25T23:32:59Z

src/vllm_router/utils.py

                return "/v1/score"
            case ModelType.transcription:
                return "/v1/audio/transcriptions"
+            case ModelType.messages:


This is mostly cosmetics at the moment, since compatible models presumably have chat defined as their type, but maybe in the future it'd make sense to support multiple types and run health checks against supported endpoints.

feat(router): add initial support for anthropic messages endpoint

0975469

Signed-off-by: Nejc Habjan <nejc.habjan@siemens.com>

gemini-code-assist bot reviewed Dec 25, 2025

View reviewed changes

nejch marked this pull request as ready for review December 25, 2025 23:33

nejch commented Dec 25, 2025

View reviewed changes

This was referenced Dec 26, 2025

[Bug]: GLM-4.7-FP8 missing beginning <think> tag vllm-project/vllm#31319

Open

fixed glm 4.7 tool call and parser vllm-project/vllm#31220

Open

[Roadmap] vLLM Production Stack roadmap for 2025 Q3 #640

Open

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

feat(router): add initial support for anthropic messages endpoint #775

feat(router): add initial support for anthropic messages endpoint #775

nejch commented Dec 25, 2025 •

edited

Loading

Uh oh!

gemini-code-assist bot commented Dec 25, 2025

Uh oh!

gemini-code-assist bot left a comment

Uh oh!

gemini-code-assist bot Dec 25, 2025

Uh oh!

nejch Dec 25, 2025

Uh oh!

nejch Dec 25, 2025

Uh oh!

nejch Dec 25, 2025

Uh oh!

nejch Dec 25, 2025

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

feat(router): add initial support for anthropic messages endpoint #775

Are you sure you want to change the base?

feat(router): add initial support for anthropic messages endpoint #775

Conversation

nejch commented Dec 25, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

PR Title and Classification

Code Quality

DCO and Signed-off-by

What to Expect for the Reviews

Uh oh!

gemini-code-assist bot commented Dec 25, 2025

Summary of Changes

Highlights

Footnotes

Uh oh!

gemini-code-assist bot left a comment

Choose a reason for hiding this comment

Code Review

Uh oh!

gemini-code-assist bot Dec 25, 2025

Choose a reason for hiding this comment

Uh oh!

nejch Dec 25, 2025

Choose a reason for hiding this comment

Uh oh!

nejch Dec 25, 2025

Choose a reason for hiding this comment

Uh oh!

nejch Dec 25, 2025

Choose a reason for hiding this comment

Uh oh!

nejch Dec 25, 2025

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

nejch commented Dec 25, 2025 •

edited

Loading