Skip to content

[MLObs] adding clarification notes about the metrics #35248

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
XG-xin merged 3 commits into from
Mar 16, 2026
Merged

[MLObs] adding clarification notes about the metrics #35248

Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
dbea19d
adding clairification notes about the metrics
XG-xin Mar 12, 2026
2d28952
remove typo newline
XG-xin Mar 12, 2026
c79a278
explain the metrics are only generated for certain keys
XG-xin Mar 13, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
6 changes: 6 additions & 0 deletions content/en/llm_observability/instrumentation/sdk.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1252,6 +1252,10 @@

## Enriching spans

<div class="alert alert-info">
The <code>metrics</code> parameter here refers to numeric values attached as attributes on individual spans — not <a href="/llm_observability/monitoring/metrics/">Datadog platform metrics</a>. For certain recognized keys such as <code>input_tokens</code>, <code>output_tokens</code>, and <code>total_tokens</code>, Datadog uses these span attributes to generate corresponding platform metrics (such as <code>ml_obs.span.llm.input.tokens</code>) for use in dashboards and monitors.

Check warning on line 1256 in content/en/llm_observability/instrumentation/sdk.md

View workflow job for this annotation

GitHub Actions / vale

Datadog.dashes 

Don't put a space before or after a dash.
</div>

{{< tabs >}}
{{% tab "Python" %}}
The SDK provides the method `LLMObs.annotate()` to enrich spans with inputs, outputs, and metadata.
Expand Down Expand Up @@ -1923,6 +1927,8 @@

If you're using automatic instrumentation, token and cost metrics appear on your spans automatically. If you're instrumenting manually, follow the guidance below.

<div class="alert alert-info">In this context, "token metrics" and "cost metrics" refer to numeric key-value pairs you attach to spans through the <code>metrics</code> parameter of the <code>LLMObs.annotate()</code> method. These are distinct from <a href="/llm_observability/monitoring/metrics/">Datadog platform LLM Observability metrics</a>. For recognized keys such as <code>input_tokens</code>, <code>output_tokens</code>, <code>input_cost</code>, and <code>output_cost</code>, Datadog uses these span attributes to generate corresponding platform metrics (such as <code>ml_obs.span.llm.input.cost</code>) for use in dashboards and monitors.</div>

Check warning on line 1930 in content/en/llm_observability/instrumentation/sdk.md

View workflow job for this annotation

GitHub Actions / vale

Datadog.words_case_insensitive 

Use 'see', 'read', or 'follow' instead of 'refer to'.

{{< tabs >}}
{{% tab "Python" %}}

Expand Down
Loading