OpenAI compatible text embedding - OpenSearch - Alibaba Cloud Documentation Center

Generates vector embeddings for one or more text inputs using the OpenSearch compatible mode embedding service.

Endpoint

POST {host}/compatible-mode/v1/embeddings

host: The endpoint for the service. The service is accessible over the Internet or a Virtual Private Cloud (VPC). To get your endpoint, see Query service endpoint.

Request parameters

Parameter	Type	Required	Description	Example
`model`	String	Yes	The service ID to use. System service IDs start with `ops-`. For supported service IDs, see List of supported services.	`ops-text-embedding-001`
`input`	Array/String	Yes	The text to embed. Accepts a single string or an array of strings. Maximum 32 inputs per request. The maximum length of each input depends on the selected model. Empty strings are not accepted.	`["Science and technology are the primary productive forces", "OpenSearch product documentation"]`

Response parameters

Parameter	Type	Description	Example
`data`	List	The embedding results.	—
`data.embedding`	Array[float]	The embedding vector for the input at the corresponding index. `ops-text-embedding-001` returns 1536 floats.	`[0.003143, 0.009750, ..., -0.017395]`
`data.index`	Int	The position of this result in the request input array, starting from `0`.	`0`
`data.object`	String	The object type. Always `embedding`.	`embedding`
`object`	String	The response type. Always `list`.	`list`
`model`	String	The service ID specified in the request.	`ops-text-embedding-001`
`usage`	Object	Token usage statistics for this request.	—
`usage.prompt_tokens`	Int	The number of tokens in the request input.	`5`
`usage.total_tokens`	Int	The total number of tokens used (`usage.prompt_tokens` + `usage.completion_tokens`).	`5`

Example

Request

curl http://xxxx-shanghai.opensearch.aliyuncs.com/compatible-mode/v1/embeddings \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer <your-api-key>" \
  -d '{
    "model": "ops-text-embedding-001",
    "input": "Search development platform"
  }'

Response

{
  "id": "9085118afe14bee71d59c4a609fcf7c0",
  "object": "list",
  "data": [
    {
      "object": "embedding",
      "embedding": [
        0.0021381378173828125,
        0.12109375,
        ...,
        -0.0389404296875
      ],
      "index": 0
    }
  ],
  "model": "ops-text-embedding-001",
  "usage": {
    "prompt_tokens": 5,
    "total_tokens": 5
  }
}

Status codes

For status code descriptions, see Status code description.