NAME
OpenSearch::Client::Core::3_0::Direct::KNN
VERSION
version 3.007004
SYNOPSIS
use OpenSearch::Client;
my $client = OpenSearch::Client->new( ... );
my $response = $client->knn-><methodname>(
valone => $value1,
valtwo => $value2
);DESCRIPTION
Vector search API
In OpenSearch, vector search functionality is provided by the k-NN plugin and Neural Search plugin. The k-NN plugin provides basic k-NN functionality, while the Neural Search plugin provides automatic embedding generation at indexing and search time.
See OpenSearch documentation for knn.
METHODS
knn->delete_model
Used to delete a particular model in the cluster.
Paths served by this method:
$resp = $client->knn->delete_model(
# path parameters
'model_id' => $model_id, # required
# Common API query string parameters
'error_trace' => $qval1, # boolean
'filter_path' => $qval2, # list
'human' => $qval3, # boolean
'pretty' => $qval4, # boolean
'source' => $qval5, # string
);OpenSearch documentation for knn.delete_model
knn->get_model
Used to retrieve information about models present in the cluster.
Paths served by this method:
$resp = $client->knn->get_model(
# path parameters
'model_id' => $model_id, # required
# Common API query string parameters
'error_trace' => $qval1, # boolean
'filter_path' => $qval2, # list
'human' => $qval3, # boolean
'pretty' => $qval4, # boolean
'source' => $qval5, # string
);OpenSearch documentation for knn.get_model
knn->search_models
Use an OpenSearch query to search for models in the index.
Paths served by this method:
$resp = $client->knn->search_models(
'body' => $body, # optional
# Endpoint specific query string parameters
'_source' => $qval1, # list
'_source_excludes' => $qval2, # list
'_source_includes' => $qval3, # list
'allow_no_indices' => $qval4, # boolean
'allow_partial_search_results' => $qval5, # boolean
'analyze_wildcard' => $qval6, # boolean
'analyzer' => $qval7, # string
'batched_reduce_size' => $qval8, # number
'ccs_minimize_roundtrips' => $qval9, # boolean
'default_operator' => $qval10, # string
'df' => $qval11, # string
'docvalue_fields' => $qval12, # list
'expand_wildcards' => $qval13, # list
'explain' => $qval14, # boolean
'from' => $qval15, # number
'ignore_throttled' => $qval16, # boolean
'ignore_unavailable' => $qval17, # boolean
'lenient' => $qval18, # boolean
'max_concurrent_shard_requests' => $qval19, # number
'pre_filter_shard_size' => $qval20, # number
'preference' => $qval21, # string
'q' => $qval22, # string
'request_cache' => $qval23, # boolean
'rest_total_hits_as_int' => $qval24, # boolean
'routing' => $qval25, # list
'scroll' => $qval26, # string
'search_type' => $qval27, # string
'seq_no_primary_term' => $qval28, # boolean
'size' => $qval29, # number
'sort' => $qval30, # list
'stats' => $qval31, # list
'stored_fields' => $qval32, # list
'suggest_field' => $qval33, # string
'suggest_mode' => $qval34, # string
'suggest_size' => $qval35, # number
'suggest_text' => $qval36, # string
'terminate_after' => $qval37, # number
'timeout' => $qval38, # string
'track_scores' => $qval39, # boolean
'track_total_hits' => $qval40, # boolean
'typed_keys' => $qval41, # boolean
'version' => $qval42, # boolean
# Common API query string parameters
'error_trace' => $qval43, # boolean
'filter_path' => $qval44, # list
'human' => $qval45, # boolean
'pretty' => $qval46, # boolean
'source' => $qval47, # string
);OpenSearch documentation for knn.search_models
knn->stats
Provides information about the current status of the k-NN plugin.
Paths served by this method:
GET /_plugins/_knn/statsGET /_plugins/_knn/stats/{stat}GET /_plugins/_knn/{node_id}/statsGET /_plugins/_knn/{node_id}/stats/{stat}
$resp = $client->knn->stats(
# path parameters
'node_id' => $node_id, # optional
'stat' => $stat, # optional
# Endpoint specific query string parameters
'timeout' => $qval1, # string
# Common API query string parameters
'error_trace' => $qval2, # boolean
'filter_path' => $qval3, # list
'human' => $qval4, # boolean
'pretty' => $qval5, # boolean
'source' => $qval6, # string
);OpenSearch documentation for knn.stats
knn->train_model
Create and train a model that can be used for initializing k-NN native library indexes during indexing.
Paths served by this method:
$resp = $client->knn->train_model(
'body' => $body, # optional
# path parameters
'model_id' => $model_id, # optional
# Endpoint specific query string parameters
'preference' => $qval1, # string
# Common API query string parameters
'error_trace' => $qval2, # boolean
'filter_path' => $qval3, # list
'human' => $qval4, # boolean
'pretty' => $qval5, # boolean
'source' => $qval6, # string
);OpenSearch documentation for knn.train_model
knn->warmup
Preloads native library files into memory, reducing initial search latency for specified indexes.
Paths served by this method:
$resp = $client->knn->warmup(
# path parameters
'index' => $index, # required
# Common API query string parameters
'error_trace' => $qval1, # boolean
'filter_path' => $qval2, # list
'human' => $qval3, # boolean
'pretty' => $qval4, # boolean
'source' => $qval5, # string
);OpenSearch documentation for knn.warmup
MANUAL
Documentation index OpenSearch::Client::Manual
HISTORY
This distribution is derived from Search::Elasticsearch version 7.714. All subsequent changes are unique to this distribution.
AUTHOR
Mark Dootson <mdootson@cpan.org> ( current maintainer )
CREDITS
OpenSearch::Client is based on Search::Elasticsearch version 7.714 by Enrico Zimuel <enrico.zimuel@elastic.co>.
COPYRIGHT AND LICENSE
Copyright (C) 2026 by Mark Dootson ( this distribution )
Copyright (C) 2021 by Elasticsearch BV ( original distribution )
This is free software, licensed under:
The Apache License, Version 2.0, January 2004