Thanks to visit codestin.com
Credit goes to github.com

Skip to content

Conversation

@aakash232
Copy link
Contributor

@aakash232 aakash232 commented Sep 21, 2025

Why are these changes needed?

This PR improves the documentation for config_list filtering by:

  • Clarifying that filters can accept both single string values and lists for models (e.g., "model": "gpt-5-nano" or "model": ["gpt-5-nano", "gpt-4.1"]).
  • Updating tag examples to show both single and multiple filtering cases consistently.
  • Incorporated feedback on filter_config docstring in openai_utils.py— now aligned with the main docs:
    • Clarified that filter_dict supports both single string and list/set inputs.
    • Added example for single-model (string) filtering.

These changes make the docs more flexible, consistent, and user-friendly.


Related issue number

Closes #1823


Screenshots (Locally Served)

image image

Checks

@aakash232
Copy link
Contributor Author

@randombet

Discussion / Follow-ups

I’m keeping this as an open discussion on whether we should maintain examples for both single and multiple models, or simplify to lists only.

Advantages of keeping both examples:

  • Clarity → avoids confusion about whether a single string is valid.
  • Flexibility → supports minimal cases (single) and batch use cases (multiple).
  • Symmetry → consistent with tag filtering behavior.

Future work:
This raises a follow-up task: updating all other notebooks to ensure filtering examples follow the same consistent style.

@randombet
Copy link
Collaborator

@aakash232 Did you verify the documented use cases for config_list is working? How did you do that?

@aakash232
Copy link
Contributor Author

aakash232 commented Sep 23, 2025

@aakash232 Did you verify the documented use cases for config_list is working? How did you do that?

@randombet
I did the local notebook setup and ran (+ logged the outputs for) the cells using both ways.
Currently, FOR A SINGLE PARAMETER, both ways are working "pass as a string" and "pass one element via list".

Based on the discussions (in github), I went ahead and supported the proposal + usage of "pass as a string" pattern.

P.S. Another solid check would be to do this in all notebooks. (Which points to the creation of a new issue/PR, as stated)

@randombet
Copy link
Collaborator

@aakash232 Please follow https://docs.ag2.ai/latest/docs/contributor-guide/pre-commit/ and fix the error.

Copy link
Collaborator

@randombet randombet left a comment

Choose a reason for hiding this comment

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

LGTM

@randombet randombet added this pull request to the merge queue Sep 26, 2025
Merged via the queue into ag2ai:main with commit b4fc5a0 Sep 26, 2025
13 checks passed
@codecov
Copy link

codecov bot commented Sep 26, 2025

Codecov Report

✅ All modified and coverable lines are covered by tests.

Files with missing lines Coverage Δ
autogen/oai/openai_utils.py 62.97% <ø> (ø)

... and 41 files with indirect coverage changes

🚀 New features to boost your workflow:
  • ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
  • 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

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.

[docs]: LLMConfig.where syntax

2 participants