Path parameters

  • task_typestring Required

    The type of the inference task that the model will perform.

    Value is sparse_embedding.

  • elser_inference_idstring Required

    The unique identifier of the inference endpoint.

application/json

Body

  • chunking_settingsobject
    Hide chunking_settings attributes Show chunking_settings attributes object
    • max_chunk_sizenumber

      The maximum size of a chunk in words. This value cannot be higher than 300 or lower than 20 (for sentence strategy) or 10 (for word strategy).

    • overlapnumber

      The number of overlapping words for chunks. It is applicable only to a word chunking strategy. This value cannot be higher than half the max_chunk_size value.

    • sentence_overlapnumber

      The number of overlapping sentences for chunks. It is applicable only for a sentence chunking strategy. It can be either 1 or 0.

    • strategystring

      The chunking strategy: sentence or word.

  • servicestring Required

    Value is elser.

  • service_settingsobject Required
    Hide service_settings attributes Show service_settings attributes object
    • adaptive_allocationsobject
      Hide adaptive_allocations attributes Show adaptive_allocations attributes object
    • num_allocationsnumber Required

      The total number of allocations this model is assigned across machine learning nodes. Increasing this value generally increases the throughput. If adaptive allocations is enabled, do not set this value because it's automatically set.

    • num_threadsnumber Required

      The number of threads used by each model allocation during inference. Increasing this value generally increases the speed per inference request. The inference process is a compute-bound process; threads_per_allocations must not exceed the number of available allocated processors per node. The value must be a power of 2. The maximum value is 32.


      If you want to optimize your ELSER endpoint for ingest, set the number of threads to 1. If you want to optimize your ELSER endpoint for search, set the number of threads to greater than 1.

Responses

  • 200 application/json
    Hide response attributes Show response attributes object
    • chunking_settingsobject
      Hide chunking_settings attributes Show chunking_settings attributes object
      • max_chunk_sizenumber

        The maximum size of a chunk in words. This value cannot be higher than 300 or lower than 20 (for sentence strategy) or 10 (for word strategy).

      • overlapnumber

        The number of overlapping words for chunks. It is applicable only to a word chunking strategy. This value cannot be higher than half the max_chunk_size value.

      • sentence_overlapnumber

        The number of overlapping sentences for chunks. It is applicable only for a sentence chunking strategy. It can be either 1 or 0.

      • strategystring

        The chunking strategy: sentence or word.

    • servicestring Required

      The service type

    • service_settingsobject Required
    • task_settingsobject
    • inference_idstring Required

      The inference Id

    • task_typestring Required

      Value is sparse_embedding.

PUT /_inference/{task_type}/{elser_inference_id}
PUT _inference/sparse_embedding/my-elser-model
{
    "service": "elser",
    "service_settings": {
        "num_allocations": 1,
        "num_threads": 1
    }
}
curl \
 --request PUT 'http://api.example.com/_inference/{task_type}/{elser_inference_id}' \
 --header "Authorization: $API_KEY" \
 --header "Content-Type: application/json" \
 --data '"{\n    \"service\": \"elser\",\n    \"service_settings\": {\n        \"num_allocations\": 1,\n        \"num_threads\": 1\n    }\n}"'
Request examples
Run `PUT _inference/sparse_embedding/my-elser-model` to create an inference endpoint that performs a `sparse_embedding` task. The request will automatically download the ELSER model if it isn't already downloaded and then deploy the model.
{
    "service": "elser",
    "service_settings": {
        "num_allocations": 1,
        "num_threads": 1
    }
}
Run `PUT _inference/sparse_embedding/my-elser-model` to create an inference endpoint that performs a `sparse_embedding` task with adaptive allocations. When adaptive allocations are enabled, the number of allocations of the model is set automatically based on the current load.
{
    "service": "elser",
    "service_settings": {
        "adaptive_allocations": {
            "enabled": true,
            "min_number_of_allocations": 3,
            "max_number_of_allocations": 10
        },
        "num_threads": 1
    }
}
Response examples (200)
A successful response when creating an ELSER inference endpoint.
{
  "inference_id": "my-elser-model",
  "task_type": "sparse_embedding",
  "service": "elser",
  "service_settings": {
    "num_allocations": 1,
    "num_threads": 1
  },
  "task_settings": {}
}