---
title: "Returns the benchmark's prompt embedding map."
method: GET
path: "/benchmark/{benchmarkId}/prompt-embedding-map"
tags: ["Benchmark", "PromptEmbeddingMap"]
---

# Returns the benchmark's prompt embedding map.

`GET /benchmark/{benchmarkId}/prompt-embedding-map`

Unpaged: the whole corpus arrives in one response because the plot needs every point at once.
 Tag strings are de-duplicated into a map-level tags vocabulary that each point indexes
 via tagIndices. Prompts without English text are not embedded, so promptCount can
 be lower than the benchmark's prompt total and the map is not a complete corpus listing.
 Responds with 204 No Content when no map has been computed yet.

## Path parameters

- `benchmarkId` string, required

## Response `200`

OK

- GetPromptEmbeddingMapEndpointOutput
  - `computedAt` string, date-time, required — When the map finished computing.
  - `promptCount` integer, required — How many prompts the map contains. Prompts without English text are not embedded, so this can be lower than the benchmark's prompt total.
  - `tags` Tag[], required — The de-duplicated tag vocabulary for the map. Each point's tagIndices are positions in this list.
    - `value` string, required
    - `category` string, nullable
  - `points` GetPromptEmbeddingMapEndpointOutputPoint[], required — One point per embedded prompt.
    - `promptId` string, required
    - `x` number, double, required
    - `y` number, double, required
    - `source` string, nullable
    - `prompt` string, nullable
    - `tagIndices` integer[], required

## Other responses

- `204` — No Content
- `400` — Bad Request
- `401` — Unauthenticated
- `403` — Forbidden

---

[API](https://skmtc.net/rapidata/apis/rapidata-api.md) · [All operations](https://skmtc.net/rapidata/apis/rapidata-api/llms.txt) · [OpenAPI document](https://skmtc-service-staging.skmtc.workers.dev/v1/apis/rapidata/rapidata-api/revisions/d779de93a5ee/schema)