Skip to content

Conversation

prane-eth
Copy link
Contributor

@prane-eth prane-eth commented Aug 21, 2025

This is a step to solve issue traceloop/openllmetry#3257

Important

Added new reasoning-related attributes to OpenLLMetry semantic conventions for OpenAI models.

  • Attributes Added:
    • gen_ai.usage.reasoning_tokens: Total reasoning tokens used in completion_tokens.
    • gen_ai.request.reasoning_effort: Reasoning effort in request (minimal, low, medium, high).
    • gen_ai.request.reasoning_summary: Reasoning summary level in request (auto, concise, detailed).
    • gen_ai.response.reasoning_effort: Actual reasoning effort used.

This description was created by Ellipsis for 32766b7. You can customize this summary. It will automatically update as commits are pushed.

Copy link

@ellipsis-dev ellipsis-dev bot left a comment

Choose a reason for hiding this comment

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

Important

Looks good to me! 👍

Reviewed everything up to 32766b7 in 37 seconds. Click for details.
  • Reviewed 15 lines of code in 1 files
  • Skipped 0 files when reviewing.
  • Skipped posting 4 draft comments. View those below.
  • Modify your settings and rules to customize what types of comments Ellipsis leaves. And don't forget to react with 👍 or 👎 to teach Ellipsis.
1. openllmetry/contributing/semantic-conventions.mdx:37
  • Draft comment:
    Clarify if reasoning_tokens are a subset of completion_tokens or calculated separately.
  • Reason this comment was not posted:
    Confidence changes required: 50% <= threshold 50% None
2. openllmetry/contributing/semantic-conventions.mdx:38
  • Draft comment:
    List the allowed enum values explicitly for reasoning_effort to aid clarity.
  • Reason this comment was not posted:
    Confidence changes required: 50% <= threshold 50% None
3. openllmetry/contributing/semantic-conventions.mdx:39
  • Draft comment:
    Consider defining permitted values for reasoning_summary to ensure consistency.
  • Reason this comment was not posted:
    Confidence changes required: 50% <= threshold 50% None
4. openllmetry/contributing/semantic-conventions.mdx:40
  • Draft comment:
    Ensure the definition of response reasoning_effort aligns with the request counterpart for consistency.
  • Reason this comment was not posted:
    Confidence changes required: 50% <= threshold 50% None

Workflow ID: wflow_JlnJybFIMYgsWyJi

You can customize Ellipsis by changing your verbosity settings, reacting with 👍 or 👎, replying to comments, or adding code review rules.

@prane-eth prane-eth changed the title Added reasoning attributes docs: add reasoning attributes Aug 21, 2025
Copy link
Member

@nirga nirga left a comment

Choose a reason for hiding this comment

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

Thanks @prane-eth!

@nirga nirga merged commit 60c8e56 into traceloop:main Aug 25, 2025
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