Text Embeddings Inference

Get Embeddings. Returns a 424 status code if the model is not an embedding model.

post/embed

Request body

dimensionsinteger nullable

The number of dimensions that the output embeddings should have. If not set, the original shape of the representation will be returned instead.

normalizeboolean

prompt_namestring nullable

The name of the prompt that should be used by for encoding. If not set, no prompt will be applied.

Must be a key in the sentence-transformers configuration prompts dictionary.

For example if prompt_name is "query" and the prompts is {"query": "query: ", ...}, then the sentence "What is the capital of France?" will be encoded as "query: What is the capital of France?" because the prompt text will be prepended before any text to encode.

truncateboolean nullable

truncation_direction'Left' | 'Right'

Example request

{
  "prompt_name": "null"
}

Response

Embeddings

number[] required

Example response

Changes

Changed in 4 of the 42 revisions of this API.3813

Sep 1, 20259988d0e8962042See the full diff
- ●
  added the new Empty enum value to the error_type response property for the response status 413
  response-property-enum-value-added
- ●
  added the new Empty enum value to the error_type response property for the response status 422
  response-property-enum-value-added
- ●
  added the new Empty enum value to the error_type response property for the response status 424
  response-property-enum-value-added
- ●
  added the new Empty enum value to the error_type response property for the response status 429
  response-property-enum-value-added
- ○
  added the new optional request property dimensions
  new-optional-request-property
- ○
  added the non-success response with the status 400
  response-non-success-status-added
Jun 28, 2024b923798e386814See the full diff
- ▲
  removed subschema #1 subschema #2 from the inputs request property oneOf list
  request-property-one-of-removed
- ○
  added the new optional request property prompt_name
  new-optional-request-property
- ○
  added the new optional request property truncation_direction
  new-optional-request-property
- ○
  the request property truncate became nullable
  request-property-became-nullable
- ○
  added InputType subschema #2 to the inputs request property oneOf list
  request-property-one-of-added
This revision also has 4 changes that name no endpoint, such as unreferenced schemas being removed. See the revision's changelog
Nov 15, 2023c4554b0fed8e1See the full diff
- ○
  added the new optional request property normalize
  new-optional-request-property
Oct 13, 20238c00768ef703246See the full diff
- ▲
  the request property truncate became not nullable
  request-property-became-not-nullable
- ▲
  removed InputType subschema #2 from the inputs request property oneOf list
  request-property-one-of-removed
- ●
  removed the request property dimensions
  request-property-removed
- ●
  removed the request property normalize
  request-property-removed
- ●
  removed the request property prompt_name
  request-property-removed
- ●
  removed the request property truncation_direction
  request-property-removed
- ○
  added subschema #1 subschema #2 to the inputs request property oneOf list
  request-property-one-of-added
- ○
  removed the non-success response with the status 400
  response-non-success-status-removed
- ○
  removed the Empty enum value from the error_type response property for the response status 413
  response-property-enum-value-removed
- ○
  removed the Empty enum value from the error_type response property for the response status 422
  response-property-enum-value-removed
- ○
  removed the Empty enum value from the error_type response property for the response status 424
  response-property-enum-value-removed
- ○
  removed the Empty enum value from the error_type response property for the response status 429
  response-property-enum-value-removed
This revision also has 30 changes that name no endpoint, such as unreferenced schemas being removed. See the revision's changelog

Full history