Generate an embedding vector

Generates an embedding vector for the provided text using the configured embedding provider (default: Google Gemini). Used by publishing services to embed entity text before indexing, and by the global search endpoint to vectorise natural language queries.

POST /search/embeddings

Generate an embedding vector

cURL

curl --request POST \
  --url https://api.example.com/search/embeddings \
  --header 'Content-Type: application/json' \
  --data '
{
  "text": "<string>",
  "model": "gemini-embedding-001"
}
'

200

{
  "embedding": [
    123
  ],
  "model": "<string>",
  "dimensions": 123
}

Body

application/json

Request body for POST /utils/embeddings.

text · string · required

Text to embed

Minimum string length: 1

model · string | null

Model override. Uses the server-configured default if omitted.

Example:

"gemini-embedding-001"

Response

200

application/json

Successful Response

Response from POST /utils/embeddings.

embedding · number[] · required

Embedding vector

model · string · required

Model used to generate the embedding

dimensions · integer · required

Number of dimensions in the embedding vector