Create a trained model vocabulary Generally available; Added in 8.0.0

PUT /_ml/trained_models/{model_id}/vocabulary

This API is supported only for natural language processing (NLP) models. The vocabulary is stored in the index as described in inference_config.*.vocabulary of the trained model definition. ##Required authorization

  • Cluster privileges: manage_ml

Path parameters

  • model_id string Required

    The unique identifier of the trained model.

application/json

Body Required

  • vocabulary array[string] Required

    The model vocabulary, which must not be empty.

  • merges array[string] Generally available; Added in 8.2.0

    The optional model merges if required by the tokenizer.

  • scores array[number] Generally available; Added in 8.9.0

    The optional vocabulary value scores if required by the tokenizer.

Responses

  • 200 application/json
    Hide response attribute Show response attribute object
    • acknowledged boolean Required

      For a successful response, this value is always true. On failure, an exception is returned instead.
PUT /_ml/trained_models/{model_id}/vocabulary 
PUT _ml/trained_models/elastic__distilbert-base-uncased-finetuned-conll03-english/vocabulary
{
  "vocabulary": [
    "[PAD]",
    "[unused0]",
    ...
  ]
}
curl \
 --request PUT 'http://api.example.com/_ml/trained_models/{model_id}/vocabulary' \
 --header "Content-Type: application/json" \
 --data '"{\n  \"vocabulary\": [\n    \"[PAD]\",\n    \"[unused0]\",\n    ...\n  ]\n}"'
Request example
An example body for a `PUT _ml/trained_models/elastic__distilbert-base-uncased-finetuned-conll03-english/vocabulary` request. 
{
  "vocabulary": [
    "[PAD]",
    "[unused0]",
    ...
  ]
}

Documentation preview

This is a preview of your version @2025-06-09 which is not yet released.

View current documentation