feat(llmobs): emit gen_ai.* attributes on APM spans

[mz1119]

Description

The APM trace UI merges gen_ai.* keys into spans client-side from the LLMObs track, so the values render but aren't indexed — you can't filter, facet, or monitor on model, provider, or token usage in APM. web-ui code ishere

This change emits a subset from tracer instead, making them real searchable APM tags: gen_ai.operation.name, gen_ai.request.model, gen_ai.provider.name, gen_ai.application.name, and gen_ai.conversation.id, plus gen_ai.usage.{input,output,total,cache_read_input,cache_write_input,reasoning_output}_tokens as metrics.

Message bodies (input, output, tool definitions, retrieval documents) stay off the APM span and continue to come from the LLMObs track.

Testing

gen_ai fields are queryable in APM if sent with changes made in the SDK. Example span

Querying that span through the APM span API shows the tags in the payload (trimmed to the relevant keys):

"custom": {
  "component": "openai",
  "gen_ai": {
    "application": { "name": "max-test" },
    "operation": { "name": "llm" },
    "provider": { "name": "openai" },
    "request": { "model": "gpt-4.1-mini-2025-04-14" },
    "usage": {
      "input_tokens": 13,
      "output_tokens": 10,
      "total_tokens": 23,
      "cache_read_input_tokens": 0,
      "cache_write_input_tokens": 0,
      "reasoning_output_tokens": 0
    }
  }
}

Unit coverage in tests/llmobs/test_llmobs_gen_ai_apm_tags.py.

[datadog-datadog-prod-us1-2]

🎉 All green!

🧪 All tests passed
❄️ No new flaky tests detected

_{This comment will be updated automatically if new data arrives.
🔗 Commit SHA: 599c4fe | Docs | View more details | Give us feedback!}

[pr-commenter]

Benchmarks

Benchmark execution time: 2026-09-11 15:54:14

Comparing candidate commit f9254a2 in PR branch max.zhang/llmobs-gen-ai-apm-tags with baseline commit 8df3fb0 in branch main.

📊 Benchmarking dashboard

Found 0 performance improvements and 3 performance regressions! Performance is the same for 590 metrics, 10 unstable metrics, 7 known flaky benchmarks, 17 flaky benchmarks without significant changes.

Explanation

This is an A/B test comparing a candidate commit's performance against that of a baseline commit. Performance changes are noted in the tables below as:

• 🟩 = significantly better candidate vs. baseline
• 🟥 = significantly worse candidate vs. baseline

We compute a confidence interval (CI) over the relative difference of means between metrics from the candidate and baseline commits, considering the baseline as the reference.

If the CI is entirely outside the configured SIGNIFICANT_IMPACT_THRESHOLD (or the deprecated UNCONFIDENCE_THRESHOLD), the change is considered significant.

Feel free to reach out to #apm-benchmarking-platform on Slack if you have any questions.

More details about the CI and significant changes

You can imagine this CI as a range of values that is likely to contain the true difference of means between the candidate and baseline commits.

CIs of the difference of means are often centered around 0%, because often changes are not that big:

---------------------------------(------|---^--------)-------------------------------->
                              -0.6%    0%  0.3%     +1.2%
                                 |          |        |
         lower bound of the CI --'          |        |
sample mean (center of the CI) -------------'        |
         upper bound of the CI ----------------------'

As described above, a change is considered significant if the CI is entirely outside the configured SIGNIFICANT_IMPACT_THRESHOLD (or the deprecated UNCONFIDENCE_THRESHOLD).

For instance, for an execution time metric, this confidence interval indicates a significantly worse performance:

----------------------------------------|---------|---(---------^---------)---------->
                                       0%        1%  1.3%      2.2%      3.1%
                                                  |   |         |         |
       significant impact threshold --------------'   |         |         |
                      lower bound of CI --------------'         |         |
       sample mean (center of the CI) --------------------------'         |
                      upper bound of CI ----------------------------------'

scenario:iastaspects-join_noaspect

• 🟥 execution_time [+11.692µs; +14.917µs] or [+7.801%; +9.953%]

scenario:iastaspects-title_aspect

• 🟥 execution_time [+52.387µs; +57.914µs] or [+18.383%; +20.322%]

scenario:samplingrules-high_match

• 🟥 execution_time [+13.968µs; +16.196µs] or [+9.061%; +10.507%]

Unstable benchmarks

These benchmarks have a confidence interval too wide to call a change; treat them as noise rather than signal.

scenario:coreapiscenario-context_with_data_listeners

• unstable execution_time [-724.867ns; +740.773ns] or [-6.574%; +6.718%]

scenario:coreapiscenario-core_dispatch_1_listener

• unstable execution_time [-36.786ns; +28.957ns] or [-5.969%; +4.698%]

scenario:coreapiscenario-core_dispatch_50_listeners

• unstable execution_time [-1694.220ns; +1585.687ns] or [-9.946%; +9.309%]

scenario:coreapiscenario-core_dispatch_exception_listeners

• unstable execution_time [-1410.067ns; +1062.994ns] or [-10.714%; +8.077%]

scenario:coreapiscenario-core_dispatch_listeners

• unstable execution_time [-324.070ns; +323.979ns] or [-8.843%; +8.840%]

scenario:coreapiscenario-core_dispatch_no_args_listeners

• unstable execution_time [-237.562ns; +266.592ns] or [-8.155%; +9.151%]

scenario:coreapiscenario-core_dispatch_with_results_1_listener

• unstable execution_time [-76.576ns; +74.987ns] or [-6.672%; +6.534%]

scenario:coreapiscenario-core_dispatch_with_results_50_listeners

• unstable execution_time [-3744.660ns; +4201.105ns] or [-9.202%; +10.324%]

scenario:coreapiscenario-core_dispatch_with_results_listeners

• unstable execution_time [-724.895ns; +876.468ns] or [-8.928%; +10.794%]

scenario:packagesupdateimporteddependencies-import_many_stdlib_cached

• unstable execution_time [-49.900µs; +62.367µs] or [-8.515%; +10.643%]

Known flaky benchmarks

These benchmarks are marked as flaky and will not trigger a failure. Modify FLAKY_BENCHMARKS_REGEX to control which benchmarks are marked as flaky.

scenario:httppropagationinject-ids_only

• 🟥 execution_time [+1.933µs; +2.059µs] or [+11.476%; +12.226%]

scenario:iastaspects-casefold_noaspect

• 🟥 execution_time [+30.200µs; +36.180µs] or [+11.936%; +14.299%]

scenario:iastaspectsospath-ospathbasename_aspect

• 🟥 execution_time [+121.133µs; +128.015µs] or [+29.507%; +31.184%]

scenario:iastaspectssplit-rsplit_aspect

• 🟥 execution_time [+13.983µs; +19.082µs] or [+9.752%; +13.308%]

scenario:span-start

• 🟥 execution_time [+1.645ms; +1.820ms] or [+11.671%; +12.916%]

scenario:telemetryaddmetric-1-count-metric-1-times

• 🟥 execution_time [+325.991ns; +368.832ns] or [+11.242%; +12.719%]

scenario:tracer-small

• 🟥 execution_time [+41.862µs; +44.419µs] or [+12.925%; +13.714%]

Known flaky benchmarks without significant changes:

• scenario:errortrackingflasksqli-baseline
• scenario:flasksimple-iast-get
• scenario:iastaspects-casefold_aspect
• scenario:iastaspects-index_aspect
• scenario:iastaspects-ljust_noaspect
• scenario:iastaspects-lower_aspect
• scenario:iastaspects-replace_aspect
• scenario:iastaspects-rstrip_aspect
• scenario:iastaspects-swapcase_aspect
• scenario:iastaspects-title_noaspect
• scenario:iastaspects-translate_aspect
• scenario:iastaspects-translate_noaspect
• scenario:iastaspects-upper_noaspect
• scenario:packagespackageforrootmodulemapping-cache_off
• scenario:packagespackageforrootmodulemapping-cache_on
• scenario:sethttpmeta-all-enabled
• scenario:telemetryaddmetric-record-100-metrics

[cit-pr-commenter-54b7da]

Codeowners resolved as

Resolved from the full PR diff against main using the target branch CODEOWNERS file.
CODEOWNERS team requests not listed below are not required by the current file set.

No remaining files require a CODEOWNERS review.

[cit-pr-commenter-54b7da]

Circular import analysis

⚠️ Existing circular imports

There are 1 circular imports that already exist on the base branch and have not been changed by this PR.

ddtrace.errortracking._handled_exceptions.bytecode_injector -> ddtrace.errortracking._handled_exceptions.callbacks -> ddtrace.errortracking._handled_exceptions.collector -> ddtrace.errortracking._handled_exceptions.bytecode_reporting -> ddtrace.errortracking._handled_exceptions.bytecode_injector

[cit-pr-commenter-54b7da]

Dependency direction analysis

⚠️ Existing dependency direction violations

There are 228 dependency direction violations that already exist on the base branch and have not been changed by this PR.

Show existing violations (showing 5 of 228 highest severity)

ddtrace.internal.tracemethods -×-> ddtrace.trace  (internal-core -> product:tracing, score=134)
ddtrace.llmobs._integrations.anthropic -×-> ddtrace.trace  (product:llmobs -> product:tracing, score=132)
ddtrace.llmobs._integrations.mcp -×-> ddtrace.trace  (product:llmobs -> product:tracing, score=132)
ddtrace.internal.opentelemetry.trace -×-> ddtrace.trace  (product:opentelemetry -> product:tracing, score=132)
ddtrace.internal.openfeature._span_enrichment -×-> ddtrace.trace  (product:openfeature -> product:tracing, score=132)

To see all violations, download the layers-base.json and layers-pr.json artifacts from this CI job and run:

uv run --script scripts/import-analysis/layers.py compare layers-base.json layers-pr.json

[chatgpt-codex-connector]

Codex Review Summary

This comment shows the latest Codex review activity on this pull request.

Review	Status	Commit	Review trigger
📝 Code Review	✅ Completed 2026-09-08T17:33:37.982750Z	3993b5c	Draft marked ready

ℹ️ About Codex in GitHub

Your team has set up Codex to 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" or "@codex security review".

Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings.

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 3993b5cb29

ℹ️ About Codex in GitHub

Your team has set up Codex to 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 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

[ncybul]

A few questions and suggestions, but overall the logic makes sense to me! What might be good is to query APM's span search API to verify that the tags are on the APM span correctly and attach the link to the APM span with the payload you searched to the PR description so it's easy to reference.

[emmettbutler]

Release note lgtm

[mz1119]

/merge

[gh-worker-devflow-routing-ef8351]

View all feedbacks in Devflow UI.

2026-09-10 18:56:47 UTC ℹ️ Start processing command /merge

---

2026-09-10 18:57:06 UTC ℹ️ MergeQueue: Pull request is not mergeable yet

It will be processed automatically as soon as GitHub reports it as mergeable. View in MergeQueue UI.

• Run /code blockers to see what is blocking it.
• Run /remove to cancel it.

---

2026-09-10 19:11:29 UTC ℹ️ MergeQueue: merge request added to the queue

The expected merge time in main is approximately 55m (p90).

---

2026-09-10 19:53:48 UTC ❌ MergeQueue: The checks failed on this merge request

Tests failed on this commit 0602375:

• serverless-system-tests-build-layer

What to do next?

• Investigate the failures and when ready, re-add your pull request to the queue!
• If your PR checks are green, try to rebase/merge. It might be because the CI run is a bit old.
• Any question, go check the FAQ.