Skip to content

embed

Endpoint Documentation: /v1/inference/embedding

This endpoint allows you to make a POST request to convert a string into a vector embedding (a List[float]).

Request

  • HTTP Method: POST
  • URL: https://api.lamini.ai/v1/inference/embedding
  • Headers:
    • Authorization: Bearer $LAMINI_API_KEY
    • Content-Type: application/json
  • Example Body (JSON):
{
    "prompt": "How old is Carl, the llama with a hat?",
    "model_name": "sentence-transformers/all-MiniLM-L6-v2"
}

Parameters:

  • prompt : str or List[str], the string to embed
  • model_name (optional): str, the name of a base or finetuned model. Default is sentence-transformers/all-MiniLM-L6-v2.

Response

If the web request is successful, you will see a response with an answer to the provided questions like below:

  • Success Status Code: 200
  • Body (JSON): 
       {
      "embedding":[0.013080810196697712,-0.05404408276081085, ... ]
   }

Otherwise, the request will return an error code, and the response json will contain specific error details like invalid token or incompatible data.

Example

Request

curl --location 'https://api.lamini.ai/v1/inference/embedding' \
   --header 'Authorization: Bearer $LAMINI_API_KEY' \
   --header 'Content-Type: application/json' \
   --data '{ "prompt": "How are you?  Rate on a scale of 1 to 5." }'

Response

{
   "embedding":[0.013080810196697712,-0.05404408276081085, ... ]
}