---
title: "List patterns topics with clustered points"
method: GET
path: "/api/v2/llm-obs/v1/topic-discovery-topics/with-cluster-points"
tags: ["LLM Observability"]
---

# List patterns topics with clustered points

`GET /api/v2/llm-obs/v1/topic-discovery-topics/with-cluster-points`

## Request

- Base URL: `https://api.datadoghq.com`
- URL: `https://api.datadoghq.com/api/v2/llm-obs/v1/topic-discovery-topics/with-cluster-points`
- Auth: API key in header `DD-API-KEY` and API key in header `DD-APPLICATION-KEY`

List the topics discovered by a patterns run, with the clustered points attached
inline to each leaf topic. When no run is specified, the most recent completed
run is used.

## Query parameters

- `config_id` string, required
- `run_id` string
- `include_metrics` boolean

## Response `200`

OK

- LLMObsPatternsTopicsWithClusteredPointsResponse: Response containing the topics, and the clustered points of their leaf topics, discovered by an LLM Observability patterns run.
  - `data` LLMObsPatternsTopicsWithClusteredPointsResponseData, required: Data object of an LLM Observability patterns topics-with-clustered-points response.
    - `attributes` LLMObsPatternsTopicsWithClusteredPointsResponseAttributes, required: Attributes of an LLM Observability patterns topics-with-clustered-points response.
      - `completed_at` string, date-time, nullable: Timestamp when the run completed. Null if the run has not completed.
      - `config_id` string, required: Identifier of the configuration that produced the run.
      - `config_snapshot` LLMObsPatternsConfigSnapshot: Snapshot of the configuration used for a patterns run.
        - `account_id` string: Integration account ID used for a bring-your-own-model run.
        - `evp_query` string: Query that selected the spans for the run.
        - `hierarchy_depth` integer: Depth of the topic hierarchy generated.
        - `integration_provider` string: Integration provider used for a bring-your-own-model run.
        - `model_name` string: Model name used for a bring-your-own-model run.
        - `num_records` integer: Maximum number of records processed for the run.
        - `sampling_ratio` number, double: Fraction of matching spans sampled for the run.
      - `created_at` string, date-time, required: Timestamp when the run was created.
      - `previous_run_id` string, required: Identifier of the run that completed immediately before this one. Empty if none.
      - `run_id` string, required: Identifier of the run that produced the topics.
      - `topics` LLMObsPatternsTopicWithClusteredPoints[], required: List of discovered topics with their clustered points.
        - `cluster_points` LLMObsPatternsClusteredPointRef[]: List of clustered points attached to a topic.
          - `duration` number, double: Duration of the source span in nanoseconds. Included only when metrics are requested.
          - `estimated_total_cost` number, double: Estimated total cost of the source span. Included only when metrics are requested.
          - `evaluation` object: Evaluation results for the source span keyed by evaluation name. Included only when metrics are requested.
          - `input_tokens` number, double: Number of input tokens of the source span. Included only when metrics are requested.
          - `output_tokens` number, double: Number of output tokens of the source span. Included only when metrics are requested.
          - `span_id` string, required: Identifier of the source span.
          - `status` string: Status of the source span. Included only when metrics are requested.
          - `total_tokens` number, double: Total number of tokens of the source span. Included only when metrics are requested.
        - `created_at` string, date-time, required: Timestamp when the topic was created.
        - `description` string, required: Description of the topic.
        - `first_seen_at` string, date-time, required: Timestamp when the topic was first seen.
        - `hierarchy_level` integer, required: Level of the topic in the hierarchy. Level 0 is a leaf topic.
        - `id` string, required: Unique identifier of the topic.
        - `is_validated` boolean, required: Whether the topic has been validated.
        - `name` string, required: Name of the topic.
        - `parent_topic_id` string, required: Identifier of the parent topic. Empty for top-level topics.
        - `point_count` integer, required: Number of data points assigned to the topic.
        - `run_id` string, required: Identifier of the run that produced the topic.
    - `id` string, required: Identifier of the run the topics belong to.
    - `type` 'get_topics_with_cluster_points_response', required: Resource type of an LLM Observability patterns topics-with-clustered-points response.

## Other responses

- `400`: Bad Request
- `401`: Unauthorized
- `403`: Forbidden
- `404`: Not Found
- `429`: Too many requests
- `500`: Internal Server Error

---

[API](https://skmtc.dev/datadog/apis/datadog-api-v2-collection.md) - [All operations](https://skmtc.dev/datadog/apis/datadog-api-v2-collection/llms.txt) - [OpenAPI document](https://skmtc.dev/datadog/apis/datadog-api-v2-collection/revisions/da68bf029e4c?raw)