Skip to content

Conversation

starpit
Copy link
Member

@starpit starpit commented Sep 19, 2025

Purpose

This PR adds two endpoints

  • /v1/query/prepare
  • /v1/query/execute

These add REST APIs around the spans core feature. They take as input a span query. This PR also adds an example query under examples/offline-inference/spans/query-{ab,ba},json. See the spans readme for an example of usage.

Test Plan

Test Result


Essential Elements of an Effective PR Description Checklist
  • The purpose of the PR, such as "Fix some issue (link existing issues this PR will resolve)".
  • The test plan, such as providing test command.
  • The test results, such as pasting the results comparison before and after, or e2e results
  • (Optional) The necessary documentation update, such as updating supported_models.md and examples for a new model.
  • (Optional) Release notes update. If your change is user facing, please update the release notes draft in the Google Doc.

Copy link

👋 Hi! Thank you for contributing to the vLLM project.

💬 Join our developer Slack at https://slack.vllm.ai to discuss your PR in #pr-reviews, coordinate on features in #feat- channels, or join special interest groups in #sig- channels.

Just a reminder: PRs would not trigger full CI run by default. Instead, it would only run fastcheck CI which starts running only a small and essential subset of CI tests to quickly catch errors.

You ask your reviewers to trigger select CI tests on top of fastcheck CI.

Once the PR is approved and ready to go, your PR reviewer(s) can run CI to test the changes comprehensively before merging.

To run CI, PR reviewers can either: Add ready label to the PR or enable auto-merge.

If you have any questions, please reach out to us on Slack at https://slack.vllm.ai.

🚀


## Span Queries

This directory contains a [span query](https://github.com/IBM/spnl#readme). To send a query, first prepare the query shape:
Copy link
Member

Choose a reason for hiding this comment

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

Should we add a preamble about how to deploy the server before running these queries against it?

Copy link
Member Author

Choose a reason for hiding this comment

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

good idea. having a turnkey approach would be good.

it also occurs to me that this README only documents the RAG use case, not the nested generation use case.

This PR adds two endpoints
- `/v1/query/prepare`
- `/v1/query/execute`

These add REST APIs around the spans core feature. They take as input a [span query](https://github.com/IBM/spnl). This PR also adds an example query under examples/offline-inference/spans/query-{ab,ba},json. See the [spans readme](examples/offline-inference/spans/README.md) for an example of usage.

Signed-off-by: Nick Mitchell <[email protected]>
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.

2 participants