traceloop / traceloop/openllmetry
🚀 Feature: Enable token usage collection by default for standalone instrumentation
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 7.4k
- Forks
- 1.1k
- Avg merge
- 8d 14h
- Merged PRs (30d)
- 2
Description
Which component is this feature for?
All Packages
🔖 Feature description
Related to https://github.com/traceloop/openllmetry/pull/3142#issuecomment-3077362491
Enable enrich_token_usage=True by default so standalone packages like opentelemetry-instrumentation-openai can collect token usage in streaming requests without explicit configuration.
Current Behavior
Standalone: OpenAIInstrumentor().instrument() → No token usage collected (default enrich_token_usage=False)
Traceloop SDK: Traceloop.init() → Token usage collected (explicitly sets enrich_token_usage=True)
Requested Behavior
Standalone instrumentation should collect token usage by default, matching traceloop-sdk behavior.
Proposed Solution
Change default parameter in OpenAIInstrumentor.init() or control this param using environment variables.
enrich_token_usage: bool = True
🎤 Why is this feature needed ?
Better observability out-of-the-box
✌️ How do you aim to achieve this?
Consistent user experience across usage patterns
🔄️ Additional Information
No response
👀 Have you spent some time to check if this feature request has been raised before?
- I checked and didn't find similar issue
Are you willing to submit PR?
Yes I am willing to submit a PR!
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Start at OpenAIInstrumentor.init() and compare standalone instrumentation with the Traceloop SDK behavior described in the issue. Update the default or environment-variable handling so enrich_token_usage is enabled for standalone use, then verify that streaming requests collect token usage without explicit configuration.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- observability
- Issue type
- Feature
- Difficulty
- 2/5
- Estimated time
- 1-3 hours
- Activity status
- Stale
- Clarity
- Mostly clear
- Newbie friendliness
- 45/100