Creates a new collection with the given parameters.

Wait for operation commit timeout in seconds.
If timeout is reached - request will return with service error.

Sharding method Default is Auto - points are distributed across all available shards Custom - points are distributed across shards according to shard key

Number of shards replicas. Default is 1 Minimum is 1

Defines how many replicas should apply the operation for us to consider it successful. Increasing this number will make the collection more resilient to inconsistencies, but will also make it fail if not enough replicas are available. Does not have any performance impact.

If true - point's payload will not be stored in memory. It will be read from the disk every time it is requested. This setting saves RAM by (slightly) increasing the response time. Note: those payload values that are involved in filtering and are indexed - remain in RAM.

Custom params for HNSW index. If none - values from service configuration file are used.

Custom params for WAL. If none - values from service configuration file are used.

Custom params for Optimizers.  If none - values from service configuration file are used.

Specify other collection to copy data from.

Quantization parameters. If none - quantization is disabled.

Create a collection

Delete a collection

Get collection details

BinaryQuantization

BinaryQuantizationConfig

CompressionRatio

Datatype

Type of internal tags, build from payload Distance function types used to compare vectors

Distance

HnswConfigDiff

Number of edges per node in the index graph. Larger the value - more accurate the search, more space required.

Number of neighbours to consider during the index building. Larger the value - more accurate the search, more time required to build the index.

Number of parallel threads used for background index building. If 0 - automatically select from 8 to 16. Best to keep between 8 and 16 to prevent likelihood of building broken/inefficient HNSW graphs. On small CPUs, less threads are used.

Store HNSW index on disk. If set to false, the index will be stored in RAM. Default: false

Custom M param for additional payload-aware HNSW links. If not set, default M will be used.

Operation for creating new collection and (optionally) specify index params

InitFrom

If used, include weight modification, which will be applied to sparse vectors at query time: None - no modification (default) Idf - inverse document frequency, based on statistics of the collection

Modifier

MultiVectorComparator

MultiVectorConfig

OptimizersConfigDiff

The minimal fraction of deleted vectors in a segment, required to perform segment optimization

The minimal number of vectors in a segment, required to perform segment optimization

Maximum size (in kilobytes) of vectors allowed for plain index, exceeding this threshold will enable vector indexing

Default value is 20,000, based on <https://github.com/google-research/google-research/blob/master/scann/docs/algorithms.md>.

To disable vector indexing, set to `0`.

Note: 1kB = 1 vector of size 256.

ProductQuantization

ProductQuantizationConfig

QuantizationConfig

ScalarQuantization

ScalarQuantizationConfig

Quantile for quantization. Expected value range in [0.5, 1.0]. If not set - use the whole range of values

If true - quantized vectors always will be stored in RAM, ignoring the config of main storage

ScalarType

ShardingMethod

SparseIndexParams

Store index on disk. If set to false, the index will be stored in RAM. Default: false

SparseIndexParamsDatatype

Params of single sparse vector data storage

SparseVectorParams

Custom params for index. If none - values from collection configuration are used.

Configures addition value modifications for sparse vectors. Default: none

SparseVectorParamsIndex

SparseVectorParamsModifier

VectorParams

Custom params for HNSW index. If none - values from collection configuration are used.

Custom params for quantization. If none - values from collection configuration are used.

If true, vectors are served from disk, improving RAM usage at the cost of latency Default: false

VectorParamsDatatype

VectorParamsHnswConfig

VectorParamsMultivectorConfig

VectorParamsQuantizationConfig

Vector params separator for single and multiple vector modes Single mode:

{ "size": 128, "distance": "Cosine" }

or multiple mode:

{ "default": { "size": 128, "distance": "Cosine" } }

VectorsConfig

WalConfigDiff

Number of WAL segments to create ahead of actually used ones

CreateCollectionHnswConfig

CreateCollectionInitFrom

Custom params for Optimizers. If none - values from service configuration file are used.

1	# Create a collection with default dense vector
2	curl -X PUT \
3	'http://localhost:6333/collections/collection_name' \
4	--header 'api-key: <api-key-value>' \
5	--header 'Content-Type: application/json' \
6	--data-raw '{
7	"vectors": {
8	"size": 384,
9	"distance": "Cosine"
10	}
11	}'
12
13	# Create a collection with named dense and sparse vectors
14	curl -X PUT \
15	'http://localhost:6333/collections/collection_name' \
16	--header 'api-key: <api-key-value>' \
17	--header 'Content-Type: application/json' \
18	--data-raw '{
19	"vectors": {
20	"dense-vector-name": {
21	"size": 1536,
22	"distance": "Cosine"
23	},
24	"sparse_vectors": {
25	"sparse-vector-name": {
26	"index": {
27	"on_disk": true
28	}
29	}
30	}
31	}
32	}'

Create a collection

Path parameters

Query parameters

Request

Response

1	{
2	"time": 0.002,
3	"status": "ok",
4	"result": true
5	}