Repository navigation
fix(ai): resolve unbounded memory growth for the vercel ai sdk llmobs plugin - #10106
Conversation
Overall package sizeSelf size: 8.74 MB Dependency sizes| name | version | self size | total size | |------|---------|-----------|------------| | import-in-the-middle | 3.4.0 | 127.33 kB | 447.04 kB | | opentracing | 0.14.7 | 194.81 kB | 194.81 kB | | dc-polyfill | 0.1.11 | 25.74 kB | 25.74 kB |🤖 This report was automatically generated by heaviest-objects-in-the-universe |
|
✅ All CI checks and tests passed. 🎉 All green!🧪 All tests passed 🎯 Code Coverage (details) 🔗 Commit SHA: ad4ab78 | Docs | View more details | Give us feedback! |
|
@codex review |
![chatgpt-codex-connector[bot]](https://avatars.githubusercontent.com/in/1144995?s=60&v=4){kind=small}
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: 055fd08f2b
ℹ️ About Codex in GitHub
Codex has been enabled to automatically review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
When you sign up for Codex through ChatGPT, Codex can also answer questions or update the PR, like "@codex address that feedback".
sabrenner
changed the title
|
@codex review |
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: 051d4af766
ℹ️ About Codex in GitHub
Codex has been enabled to automatically review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
When you sign up for Codex through ChatGPT, Codex can also answer questions or update the PR, like "@codex address that feedback".
|
@codex review |
BenchmarksBenchmark execution time: 2026-09-03 13:36:15 Comparing candidate commit ad4ab78 in PR branch Found 0 performance improvements and 0 performance regressions! Performance is the same for 2311 metrics, 17 unstable metrics.
|
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: 8179aa72b1
ℹ️ About Codex in GitHub
Codex has been enabled to automatically review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
When you sign up for Codex through ChatGPT, Codex can also answer questions or update the PR, like "@codex address that feedback".
|
@codex review |
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: 684871c53a
ℹ️ About Codex in GitHub
Codex has been enabled to automatically review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
When you sign up for Codex through ChatGPT, Codex can also answer questions or update the PR, like "@codex address that feedback".
|
@codex review |
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: a173007426
ℹ️ About Codex in GitHub
Codex has been enabled to automatically review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
When you sign up for Codex through ChatGPT, Codex can also answer questions or update the PR, like "@codex address that feedback".
|
@codex review |
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: e71ff094a8
ℹ️ About Codex in GitHub
Codex has been enabled to automatically review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
When you sign up for Codex through ChatGPT, Codex can also answer questions or update the PR, like "@codex address that feedback".
|
@codex review |
|
Codex Review: Didn't find any major issues. Keep it up! Reviewed commit: ℹ️ About Codex in GitHubCodex has been enabled to automatically review pull requests in this repo. Reviews are triggered when you
If Codex has suggestions, it will comment; otherwise it will react with 👍. When you sign up for Codex through ChatGPT, Codex can also answer questions or update the PR, like "@codex address that feedback". |
sabrenner
changed the title
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: e71ff094a8
ℹ️ About Codex in GitHub
Codex has been enabled to automatically review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
When you sign up for Codex through ChatGPT, Codex can also answer questions or update the PR, like "@codex address that feedback".
BridgeAR
left a comment
BridgeAR
left a comment
There was a problem hiding this comment.
Fixing a bug with removing code ❤️
LGTM
* cap tool emission to v4 * remove tool ids if they can be resolved without toolCallIdsToName * regression test * test more ranges in regression test * update test implementation, only test v5 * always delete tool calls id to name once consumed * better availableTools bounding * stringify raw tool json from destructured components * array-based tools were never a thing
* cap tool emission to v4 * remove tool ids if they can be resolved without toolCallIdsToName * regression test * test more ranges in regression test * update test implementation, only test v5 * always delete tool calls id to name once consumed * better availableTools bounding * stringify raw tool json from destructured components * array-based tools were never a thing
* cap tool emission to v4 * remove tool ids if they can be resolved without toolCallIdsToName * regression test * test more ranges in regression test * update test implementation, only test v5 * always delete tool calls id to name once consumed * better availableTools bounding * stringify raw tool json from destructured components * array-based tools were never a thing
* cap tool emission to v4 * remove tool ids if they can be resolved without toolCallIdsToName * regression test * test more ranges in regression test * update test implementation, only test v5 * always delete tool calls id to name once consumed * better availableTools bounding * stringify raw tool json from destructured components * array-based tools were never a thing
2 participants
What does this PR do?
Entirely removes the tool tracking in the Datadog-telemetry based plugin for the Vercel AI SDK. This is because, when initially implementing, I had thought the v4 SDK required an array of tools with an ID passed in. This was a self-hallucination over a year ago, and this was never the case. All the special handling for that can safely be removed, removing any memory overhead associated with in.
Motivation
Fix memory issues associated with the AI SDK plugin. Fixes #9276
Additional Notes
Majority LOC for new cassette fixtures added for proper tool names.