Class InferenceService

java.lang.Object
org.elasticsearch.xpack.esql.inference.InferenceService

public class InferenceService extends Object
  • Field Details

    • ESQL_PRODUCT_USE_CASE

      public static final String ESQL_PRODUCT_USE_CASE
      Value of the X-Elastic-Product-Use-Case header for ESQL requests.
      See Also:
  • Constructor Details

    • InferenceService

      public InferenceService(Client client, ClusterService clusterService)
      Creates a new inference service with the given client.
      Parameters:
      client - the Elasticsearch client for inference operations
      clusterService - used to read and update inference settings
  • Method Details

    • inferenceSettings

      public InferenceSettings inferenceSettings()
      Returns the settings for ES|QL inference features.
      Returns:
      the inference settings
    • resolveInferenceIds

      public void resolveInferenceIds(List<String> inferenceIds, ActionListener<InferenceResolution> listener)
      Resolves a list of inference deployment IDs to their metadata.
      Parameters:
      inferenceIds - List of inference deployment IDs to resolve
      listener - Callback to receive the resolution results
    • executeInference

      public void executeInference(InferenceAction.Request request, ActionListener<InferenceAction.Response> listener)
      Executes a plain inference request.
      Parameters:
      request - the inference request to execute
      listener - the listener to notify upon completion
    • executeEmbeddingInference

      public void executeEmbeddingInference(EmbeddingAction.Request request, ActionListener<InferenceAction.Response> listener)
      Executes an embedding inference request.
      Parameters:
      request - the embedding request to execute
      listener - the listener to notify upon completion
    • executeRerankInference

      public void executeRerankInference(RerankAction.Request request, ActionListener<InferenceAction.Response> listener)
      Executes a rerank inference request.
      Parameters:
      request - the rerank request to execute
      listener - the listener to notify upon completion
    • threadPool

      public ThreadPool threadPool()
    • threadContext

      public ThreadContext threadContext()