---
title: "List score samples for a target"
method: GET
path: "/v1/benchmark/target/{targetKey}/score-sample"
tags: ["Benchmark"]
---

# List score samples for a target

`GET /v1/benchmark/target/{targetKey}/score-sample`

## Request

- Base URL: `https://api.roark.ai`
- URL: `https://api.roark.ai/v1/benchmark/target/{targetKey}/score-sample`
- Auth: HTTP bearer

Returns the individual scored calls behind a target’s aggregate cells, each with the scorer’s rationale — the "why is this number what it is" read. Filter to one cell with `conditionKey` and `metricKey`. Metrics computed without a rationale (latencies, counts) contribute no samples. Returns reasoning only: no transcript and no audio.

## Path parameters

- `targetKey` string, required

## Query parameters

- `suite` string, required
- `suiteVersion` string
- `conditionKey` string
- `metricKey` string
- `limit` integer: Maximum number of records to return (default: 20, max: 100)
- `offset` integer: Pagination offset

## Response `200`

One page of score samples

- object
  - `data` BenchmarkScoreSample[], required
    - `conditionKey` string, required
    - `metricKey` string, required
    - `sourceCallId` string, required: Opaque id of the call this single score came from. Groups samples that share a call; it is not resolvable through this API.
    - `numericValue` number, nullable, required
    - `booleanValue` boolean, nullable, required
    - `reasoning` string, required: The scorer's rationale for this individual score.
  - `target` BenchmarkTarget, required: A model or stack with a published result on a benchmark suite
    - `targetKey` string, required: Stable identity of the model/stack across sweeps. Cite this, and pass it to the target endpoints.
    - `targetName` string, required: Human-readable name of the model/stack.
    - `suite` string, required
    - `suiteVersion` string, required
    - `publicationId` string, uuid, required: Id of this one sweep. Changes every time the target is re-published, so it is a render key, not an identity.
    - `iterations` integer, required: Calls the sweep requested per condition.
    - `sampleCallCount` integer, required: Executed calls that contributed at least one score to these numbers.
    - `publishedAt` string, required: When this sweep was published (ISO-8601), i.e. data freshness.
    - `supersededAt` string, nullable, required: When a later sweep replaced this one. Null means it is the current generation.
    - `isCurrent` boolean, required: Whether these are the numbers Roark publishes for this target today.
  - `pagination` BenchmarkPagination, required
    - `limit` integer, required
    - `offset` integer, required
    - `total` integer, required: Total matching records, ignoring this page.
    - `hasMore` boolean, required

## Other responses

- `400`: Bad Request
- `401`: Unauthorized
- `404`: Not Found
- `500`: Internal Server Error

## Changes

> 188 revisions in range; 1 not diffed.

- **2026-09-15** `38989bd64cb3`: 1 info
  - endpoint added

[Operation history](https://skmtc.dev/roarkhq/apis/roark-analytics-api/changes/v1/benchmark/target/:targetKey/score-sample/get.md)

---

[API](https://skmtc.dev/roarkhq/apis/roark-analytics-api.md) - [All operations](https://skmtc.dev/roarkhq/apis/roark-analytics-api/llms.txt) - [OpenAPI document](https://skmtc.dev/roarkhq/apis/roark-analytics-api/revisions/efc3377249d1?raw)