Retrieves the closest points in batches based on vector similarity and given filtering conditions.

Define read consistency guarantees for the operation

If set, overrides global timeout for this request. Unit is seconds.

Search batch points

Search point groups

Search points

AnyVariants

Condition

DatetimeRange

Document

Text of the document This field will be used as input for the embedding model

Name of the model used to generate the vector List of available models depends on a provider

Type, used for specifying point ID in user interface

ExtendedPointId

All possible payload filtering conditions

FieldCondition

Check if point has field with a given value

Check if points value lies in a given range

Check if points geo location lies in a given area

Check if geo point is within a given radius

Check if geo point is within a given polygon

FieldConditionGeoBoundingBox

FieldConditionGeoPolygon

FieldConditionGeoRadius

FieldConditionMatch

FieldConditionRange

FieldConditionValuesCount

Filter

At least one of those conditions should match

At least minimum amount of given conditions should match

FilterMinShould

FilterMust

FilterMustNot

FilterShould

GeoBoundingBox

Ordered sequence of GeoPoints representing the line

GeoLineString

GeoPoint

GeoPolygon

Interior lines (if present) bound holes within the surface each GeoLineString must consist of a minimum of 4 points, and the first and last points must be the same.

GeoRadius

HasIdCondition

Select points with empty payload for a specified field

IsEmptyCondition

Select points with null payload for a specified field

IsNullCondition

Match

MatchAny

Should have at least one value not matching the any given values

MatchExcept

MatchText

MatchValue

MinShould

NamedSparseVector

NamedVector

Vector data separator for named and unnamed modes Unnamed mode:

{ "vector": [1.0, 2.0, 3.0] }

or named mode:

{ "vector": { "vector": [1.0, 2.0, 3.0], "name": "image-embeddings" } }

NamedVectorStruct

Select points with payload for a specified nested field

Nested

NestedCondition

OrderValue

Payload

PayloadField

PayloadSelector

PayloadSelectorExclude

Exclude this fields from returning payload

PayloadSelectorInclude

QuantizationSearchParams

If true, quantized vectors are ignored. Default is false.

If true, use original vectors to re-score top-k results. Might require more time in case if original vectors are stored on disk. If not set, qdrant decides automatically apply rescoring or not.

Range

RangeInterface

ReadConsistency

ReadConsistencyType

ScoredPoint

Points vector distance to the query vector

ScoredPointOrderValue

ScoredPointPayload

ScoredPointShardKey

ScoredPointVector

SearchParams

Params relevant to HNSW index Size of the beam in a beam-search. Larger the value - more accurate the result, more time required for search.

Search without approximation. If set to true, search may run long but with exact results.

If enabled, the engine will only perform search among indexed or small segments. Using this option prevents slow searches in case of delayed index, but does not guarantee that all uploaded vectors will be included in search results

SearchParamsQuantization

Search request. Holds all conditions and parameters for the search of most similar points by vector similarity given the filtering restrictions.

SearchRequest

Specify in which shards to look for the points, if not specified - look in all shards

Look only for points which satisfies this conditions

Offset of the first result to return. May be used to paginate results. Note: large offset values may cause performance issues.

Select which payload to return with the response. Default is false.

Options for specifying which vectors to include into response. Default is false.

Define a minimal score threshold for the result. If defined, less similar results will not be returned. Score of the returned result might be higher or smaller than the threshold depending on the Distance function used. E.g. for cosine similarity only higher scores will be returned.

1	curl -X POST \
2	'http://localhost:6333/collections/collection_name/points/search/batch' \
3	--header 'api-key: <api-key-value>' \
4	--header 'Content-Type: application/json' \
5	--data-raw '{
6	"searches": [
7	{
8	"vector": [
9	0.2,
10	0.1,
11	0.9,
12	0.7
13	],
14	"limit": 3
15	},
16	{
17	"vector": [
18	0.5,
19	0.3,
20	0.2,
21	0.3
22	],
23	"filter": {
24	"must": [
25	{
26	"key": "city",
27	"match": {
28	"value": "London"
29	}
30	}
31	]
32	},
33	"limit": 3
34	}
35	]
36	}'

1	{
2	"time": 0.002,
3	"status": "ok",
4	"result": [
5	[
6	{
7	"id": 42,
8	"version": 3,
9	"score": 0.75,
10	"payload": {
11	"city": "London",
12	"color": "green"
13	},
14	"vector": [
15	0.875,
16	0.140625,
17	0.897599995136261
18	],
19	"shard_key": "region_1",
20	"order_value": 42
21	}
22	]
23	]
24	}

Search batch points

Path parameters

Query parameters

Request

Response