Clear trained model deployment cache

POST /_ml/trained_models/{model_id}/deployment/cache/_clear

Auth

Cache will be cleared on all nodes where the trained model is assigned. A trained model deployment may have an inference cache enabled. As requests are handled by each allocated node, their responses may be cached on that individual node. Calling this API clears the caches without restarting the deployment.

Required privileges

Cluster privileges manage_ml

Path parameters

model_id string required: The unique identifier of the trained model.

Response

application/json

cleared boolean required

View as Markdown

Examples

Request Console 1 / 8 languages

Console Kibana Dev Tools

POST _ml/trained_models/elastic__distilbert-base-uncased-finetuned-conll03-english/deployment/cache/_clear

JavaScript Elasticsearch JavaScript Client

const response = await client.ml.clearTrainedModelDeploymentCache({
  model_id: "elastic__distilbert-base-uncased-finetuned-conll03-english",
});

Python Elasticsearch Python Client

resp = client.ml.clear_trained_model_deployment_cache(
    model_id="elastic__distilbert-base-uncased-finetuned-conll03-english",
)

curl Shell

curl -X POST "$ELASTICSEARCH_URL/_ml/trained_models/elastic__distilbert-base-uncased-finetuned-conll03-english/deployment/cache/_clear" \
  -H "Authorization: ApiKey $ELASTIC_API_KEY"

Java Elasticsearch Java Client

client.ml().clearTrainedModelDeploymentCache(c -> c
    .modelId("elastic__distilbert-base-uncased-finetuned-conll03-english")
);

C# Elasticsearch .NET Client

var response = await client.MachineLearning.ClearTrainedModelDeploymentCacheAsync(modelId: "elastic__distilbert-base-uncased-finetuned-conll03-english");

PHP Elasticsearch PHP Client

$resp = $client->ml()->clearTrainedModelDeploymentCache([
    "model_id" => "elastic__distilbert-base-uncased-finetuned-conll03-english",
]);

Ruby Elasticsearch Ruby Client

response = client.ml.clear_trained_model_deployment_cache(
  model_id: "elastic__distilbert-base-uncased-finetuned-conll03-english"
)

Response

A successful response when clearing the inference cache.

{
  "cleared": true
}