elastic/elasticsearch-py v9.3.0

9.3.0

on GitHub

Enhancements

• Add pack_dense_vector helper function to pack dense vectors for efficient uploading (#3219)
• New and updated ES|QL functions in the ES|QL query builder for 9.3 and Serverless (#3266)

API

• Added cat.circuit_breaker API
• Added experimental esql.get_view, esql.put_view and esql.delete_view APIs
• Added experimental indices.get_sample_configuration, indices.put_sample_configuration, indices.delete_sample_configuration, indices.get_all_sample_configuration, indices.get_sample, indices.get_sample_stats APIs
• Added inference.put_groq, inference.put_openshift_ai, inference.put_nvidia APIs
• Added downsampling_method argument to indices.put_data_lifecycle API
• Added return_documents and top_n arguments to inference.rerank API
• Added close_job argument to ml.stop_datafeed API
• Added certificate_identity to security.create_cross_cluster_api_key and security.update_cross_cluster_api_key APIs

Serverless-specific

• Added project_routing argument to project.tags API

DSL

• Added NumpyDenseVector field, with support for dense vectors based on numpy arrays (#3218)
• Added ExponentialHistogram field
• Added time_series_metric argument to Histogram field
• Added on_disk_rescore argument from DenseVectorIndexOptions type
• Added slices argument to UpdateByQueryResponse type