Log in Start Free

Administration

Qdrant exposes administration tools which enable to modify at runtime the behavior of a qdrant instance without changing its configuration manually.

Recovery mode

Available as of v1.2.0

Recovery mode can help in situations where Qdrant fails to start repeatedly. When starting in recovery mode, Qdrant only loads collection metadata to prevent going out of memory. This allows you to resolve out of memory situations, for example, by deleting a collection. After resolving Qdrant can be restarted normally to continue operation.

In recovery mode, collection operations are limited to deleting a collection. That is because only collection metadata is loaded during recovery.

To enable recovery mode with the Qdrant Docker image you must set the environment variable QDRANT_ALLOW_RECOVERY_MODE=true. The container will try to start normally first, and restarts in recovery mode if initialisation fails due to an out of memory error. This behavior is disabled by default.

If using a Qdrant binary, recovery mode can be enabled by setting a recovery message in an environment variable, such as QDRANT__STORAGE__RECOVERY_MODE="My recovery message".

Strict mode

Available as of v1.13.0

Strict mode is a feature to restrict certain type of operations on a collection in order to protect the Qdrant cluster.

The goal is to prevent inefficient usage patterns that could overload the system.

Strict mode ensures a more predictable and responsive service when you do not have control over the queries that are being executed.

Upon crossing a limit, the server will return a client side error with the information about the limit that was crossed.

The strict_mode_config can be enabled when creating a new collection, see schema definitions for all the available strict_mode_config parameters.

As part of the config, the enabled field act as a toggle to enable or disable the strict mode dynamically.

It is possible to raise the default limits and/or disable strict mode entirely. Though, in order to ensure a stable cluster we strongly recommend to keep strict mode enabled using its default configuration. For disabling strict mode on an existing collection use:

PUT /collections/{collection_name}
{
    "strict_mode_config": {
        "enabled": false
    }
}

curl -X PUT http://localhost:6333/collections/{collection_name} \
  -H 'Content-Type: application/json' \
  --data-raw '{
    "strict_mode_config": {
      "enabled":" false
    }
  }'

from qdrant_client import QdrantClient, models

client = QdrantClient(url="http://localhost:6333")

client.create_collection(
    collection_name="{collection_name}",
    strict_mode_config=models.StrictModeConfig(enabled=False),
)

import { QdrantClient } from "@qdrant/js-client-rest";

const client = new QdrantClient({ host: "localhost", port: 6333 });

client.createCollection("{collection_name}", {
  strict_mode_config: {
    enabled: false,
  },
});

use qdrant_client::Qdrant;
use qdrant_client::qdrant::{CreateCollectionBuilder, StrictModeConfigBuilder};

let client = Qdrant::from_url("http://localhost:6334").build()?;

client
    .create_collection(
        CreateCollectionBuilder::new("{collection_name}")
            .strict_mode_config(StrictModeConfigBuilder::default().enabled(false)),
    )
    .await?;

import io.qdrant.client.QdrantClient;
import io.qdrant.client.QdrantGrpcClient;
import io.qdrant.client.grpc.Collections.CreateCollection;
import io.qdrant.client.grpc.Collections.StrictModeConfig;

QdrantClient client =
    new QdrantClient(QdrantGrpcClient.newBuilder("localhost", 6334, false).build());

client
    .createCollectionAsync(
        CreateCollection.newBuilder()
            .setCollectionName("{collection_name}")
            .setStrictModeConfig(
                StrictModeConfig.newBuilder().setEnabled(false).build())
            .build())
    .get();

using Qdrant.Client;
using Qdrant.Client.Grpc;

var client = new QdrantClient("localhost", 6334);

await client.CreateCollectionAsync(
  collectionName: "{collection_name}",
  strictModeConfig: new StrictModeConfig { Enabled = false }
);

import (
  "context"

  "github.com/qdrant/go-client/qdrant"
)

client, err := qdrant.NewClient(&qdrant.Config{
  Host: "localhost",
  Port: 6334,
})

client.CreateCollection(context.Background(), &qdrant.CreateCollection{
  CollectionName: "{collection_name}",
  StrictModeConfig: &qdrant.StrictModeConfig{
    Enabled: qdrant.PtrOf(false),
  },
})

Disable retrieving via non indexed payload

Setting unindexed_filtering_retrieve to false prevents retrieving points by filtering on a non indexed payload key which can be very slow.

PUT /collections/{collection_name}
{
    "strict_mode_config": {
        "enabled": true,
        "unindexed_filtering_retrieve": false
    }
}

curl -X PUT http://localhost:6333/collections/{collection_name} \
  -H 'Content-Type: application/json' \
  --data-raw '{
    "strict_mode_config": {
      "enabled":" true,
      "unindexed_filtering_retrieve": false
    }
  }'

from qdrant_client import QdrantClient, models

client = QdrantClient(url="http://localhost:6333")

client.create_collection(
    collection_name="{collection_name}",
    strict_mode_config=models.StrictModeConfig(enabled=True, unindexed_filtering_retrieve=False),
)

import { QdrantClient } from "@qdrant/js-client-rest";

const client = new QdrantClient({ host: "localhost", port: 6333 });

client.createCollection("{collection_name}", {
  strict_mode_config: {
    enabled: true,
    unindexed_filtering_retrieve: false,
  },
});

use qdrant_client::Qdrant;
use qdrant_client::qdrant::{CreateCollectionBuilder, StrictModeConfigBuilder};

let client = Qdrant::from_url("http://localhost:6334").build()?;

client
    .create_collection(
        CreateCollectionBuilder::new("{collection_name}")
            .strict_mode_config(StrictModeConfigBuilder::default().enabled(true).unindexed_filtering_retrieve(false)),
    )
    .await?;

import io.qdrant.client.QdrantClient;
import io.qdrant.client.QdrantGrpcClient;
import io.qdrant.client.grpc.Collections.CreateCollection;
import io.qdrant.client.grpc.Collections.StrictModeConfig;

QdrantClient client =
    new QdrantClient(QdrantGrpcClient.newBuilder("localhost", 6334, false).build());

client
    .createCollectionAsync(
        CreateCollection.newBuilder()
            .setCollectionName("{collection_name}")
            .setStrictModeConfig(
                StrictModeConfig.newBuilder().setEnabled(true).setUnindexedFilteringRetrieve(false).build())
            .build())
    .get();

using Qdrant.Client;
using Qdrant.Client.Grpc;

var client = new QdrantClient("localhost", 6334);

await client.CreateCollectionAsync(
  collectionName: "{collection_name}",
  strictModeConfig: new StrictModeConfig { Enabled = true, UnindexedFilteringRetrieve = false }
);

import (
  "context"

  "github.com/qdrant/go-client/qdrant"
)

client, err := qdrant.NewClient(&qdrant.Config{
  Host: "localhost",
  Port: 6334,
})

client.CreateCollection(context.Background(), &qdrant.CreateCollection{
  CollectionName: "{collection_name}",
  StrictModeConfig: &qdrant.StrictModeConfig{
    Enabled: qdrant.PtrOf(true),
    UnindexedFilteringRetrieve: qdrant.PtrOf(false),
  },
})

Or turn it off later on an existing collection through the collection update API.

PUT /collections/{collection_name}
{
    "strict_mode_config": {
        "enabled": true,
        "unindexed_filtering_retrieve": true
    }
}

curl -X PUT http://localhost:6333/collections/{collection_name} \
  -H 'Content-Type: application/json' \
  --data-raw '{
    "strict_mode_config": {
      "enabled":" true,
      "unindexed_filtering_retrieve": true
    }
  }'

from qdrant_client import QdrantClient, models

client = QdrantClient(url="http://localhost:6333")

client.create_collection(
    collection_name="{collection_name}",
    strict_mode_config=models.StrictModeConfig(enabled=True, unindexed_filtering_retrieve=True),
)

import { QdrantClient } from "@qdrant/js-client-rest";

const client = new QdrantClient({ host: "localhost", port: 6333 });

client.createCollection("{collection_name}", {
  strict_mode_config: {
    enabled: true,
    unindexed_filtering_retrieve: true,
  },
});

use qdrant_client::Qdrant;
use qdrant_client::qdrant::{CreateCollectionBuilder, StrictModeConfigBuilder};

let client = Qdrant::from_url("http://localhost:6334").build()?;

client
    .create_collection(
        CreateCollectionBuilder::new("{collection_name}")
            .strict_mode_config(StrictModeConfigBuilder::default().enabled(true).unindexed_filtering_retrieve(true)),
    )
    .await?;

import io.qdrant.client.QdrantClient;
import io.qdrant.client.QdrantGrpcClient;
import io.qdrant.client.grpc.Collections.CreateCollection;
import io.qdrant.client.grpc.Collections.StrictModeConfig;

QdrantClient client =
    new QdrantClient(QdrantGrpcClient.newBuilder("localhost", 6334, false).build());

client
    .createCollectionAsync(
        CreateCollection.newBuilder()
            .setCollectionName("{collection_name}")
            .setStrictModeConfig(
                StrictModeConfig.newBuilder().setEnabled(true).setUnindexedFilteringRetrieve(true).build())
            .build())
    .get();

using Qdrant.Client;
using Qdrant.Client.Grpc;

var client = new QdrantClient("localhost", 6334);

await client.CreateCollectionAsync(
  collectionName: "{collection_name}",
  strictModeConfig: new StrictModeConfig { Enabled = true, UnindexedFilteringRetrieve = true }
);

import (
  "context"

  "github.com/qdrant/go-client/qdrant"
)

client, err := qdrant.NewClient(&qdrant.Config{
  Host: "localhost",
  Port: 6334,
})

client.CreateCollection(context.Background(), &qdrant.CreateCollection{
  CollectionName: "{collection_name}",
  StrictModeConfig: &qdrant.StrictModeConfig{
    Enabled: qdrant.PtrOf(true),
    UnindexedFilteringRetrieve: qdrant.PtrOf(true),
  },
})

Disable updating via non indexed payload

Setting unindexed_filtering_update to false prevents updating points by filtering on a non indexed payload key which can be very slow.

PUT /collections/{collection_name}
{
    "strict_mode_config": {
        "enabled": true,
        "unindexed_filtering_update": false
    }
}

curl -X PUT http://localhost:6333/collections/{collection_name} \
  -H 'Content-Type: application/json' \
  --data-raw '{
    "strict_mode_config": {
      "enabled":" true,
      "unindexed_filtering_update": false
    }
  }'

from qdrant_client import QdrantClient, models

client = QdrantClient(url="http://localhost:6333")

client.create_collection(
    collection_name="{collection_name}",
    strict_mode_config=models.StrictModeConfig(enabled=True, unindexed_filtering_update=False),
)

import { QdrantClient } from "@qdrant/js-client-rest";

const client = new QdrantClient({ host: "localhost", port: 6333 });

client.createCollection("{collection_name}", {
  strict_mode_config: {
    enabled: true,
    unindexed_filtering_update: false,
  },
});

use qdrant_client::Qdrant;
use qdrant_client::qdrant::{CreateCollectionBuilder, StrictModeConfigBuilder};

let client = Qdrant::from_url("http://localhost:6334").build()?;

client
    .create_collection(
        CreateCollectionBuilder::new("{collection_name}")
            .strict_mode_config(StrictModeConfigBuilder::default().enabled(true).unindexed_filtering_update(false)),
    )
    .await?;

import io.qdrant.client.QdrantClient;
import io.qdrant.client.QdrantGrpcClient;
import io.qdrant.client.grpc.Collections.CreateCollection;
import io.qdrant.client.grpc.Collections.StrictModeConfig;

QdrantClient client =
    new QdrantClient(QdrantGrpcClient.newBuilder("localhost", 6334, false).build());

client
    .createCollectionAsync(
        CreateCollection.newBuilder()
            .setCollectionName("{collection_name}")
            .setStrictModeConfig(
                StrictModeConfig.newBuilder().setEnabled(true).setUnindexedFilteringUpdate(false).build())
            .build())
    .get();

using Qdrant.Client;
using Qdrant.Client.Grpc;

var client = new QdrantClient("localhost", 6334);

await client.CreateCollectionAsync(
  collectionName: "{collection_name}",
  strictModeConfig: new StrictModeConfig { Enabled = true, UnindexedFilteringUpdate = false }
);

import (
  "context"

  "github.com/qdrant/go-client/qdrant"
)

client, err := qdrant.NewClient(&qdrant.Config{
  Host: "localhost",
  Port: 6334,
})

client.CreateCollection(context.Background(), &qdrant.CreateCollection{
  CollectionName: "{collection_name}",
  StrictModeConfig: &qdrant.StrictModeConfig{
    Enabled: qdrant.PtrOf(true),
    UnindexedFilteringUpdate: qdrant.PtrOf(false),
  },
})

Maximum number of payload index count

Setting max_payload_index_count caps the maximum number of payload index that can exist on a collection.

PUT /collections/{collection_name}
{
    "strict_mode_config": {
        "enabled": true,
        "max_payload_index_count": 10
    }
}

curl -X PUT http://localhost:6333/collections/{collection_name} \
  -H 'Content-Type: application/json' \
  --data-raw '{
    "strict_mode_config": {
      "enabled":" true,
      "max_payload_index_count": 10
    }
  }'

from qdrant_client import QdrantClient, models

client = QdrantClient(url="http://localhost:6333")

client.create_collection(
    collection_name="{collection_name}",
    strict_mode_config=models.StrictModeConfig(enabled=True, max_payload_index_count=10),
)

import { QdrantClient } from "@qdrant/js-client-rest";

const client = new QdrantClient({ host: "localhost", port: 6333 });

client.createCollection("{collection_name}", {
  strict_mode_config: {
    enabled: true,
    max_payload_index_count: 10,
  },
});

use qdrant_client::Qdrant;
use qdrant_client::qdrant::{CreateCollectionBuilder, StrictModeConfigBuilder};

let client = Qdrant::from_url("http://localhost:6334").build()?;

client
    .create_collection(
        CreateCollectionBuilder::new("{collection_name}")
            .strict_mode_config(StrictModeConfigBuilder::default().enabled(true).max_payload_index_count(10)),
    )
    .await?;

import io.qdrant.client.QdrantClient;
import io.qdrant.client.QdrantGrpcClient;
import io.qdrant.client.grpc.Collections.CreateCollection;
import io.qdrant.client.grpc.Collections.StrictModeConfig;

QdrantClient client =
    new QdrantClient(QdrantGrpcClient.newBuilder("localhost", 6334, false).build());

client
    .createCollectionAsync(
        CreateCollection.newBuilder()
            .setCollectionName("{collection_name}")
            .setStrictModeConfig(
                StrictModeConfig.newBuilder().setEnabled(true).setMaxPayloadIndexCount(10).build())
            .build())
    .get();

using Qdrant.Client;
using Qdrant.Client.Grpc;

var client = new QdrantClient("localhost", 6334);

await client.CreateCollectionAsync(
  collectionName: "{collection_name}",
  strictModeConfig: new StrictModeConfig { Enabled = true, MaxPayloadIndexCount = 10 }
);

import (
  "context"

  "github.com/qdrant/go-client/qdrant"
)

client, err := qdrant.NewClient(&qdrant.Config{
  Host: "localhost",
  Port: 6334,
})

client.CreateCollection(context.Background(), &qdrant.CreateCollection{
  CollectionName: "{collection_name}",
  StrictModeConfig: &qdrant.StrictModeConfig{
    Enabled: qdrant.PtrOf(true),
    MaxPayloadIndexCount: qdrant.PtrOf(uint64(10)),
  },
})

Maximum query limit parameter

Retrieving large result set is expensive.

Setting max_query_limit caps the maximum number of points that can be retrieved in a single query.

PUT /collections/{collection_name}
{
    "strict_mode_config": {
        "enabled": true,
        "max_query_limit": 10
    }
}

curl -X PUT http://localhost:6333/collections/{collection_name} \
  -H 'Content-Type: application/json' \
  --data-raw '{
    "strict_mode_config": {
      "enabled":" true,
      "max_query_limit": 10
    }
  }'

from qdrant_client import QdrantClient, models

client = QdrantClient(url="http://localhost:6333")

client.create_collection(
    collection_name="{collection_name}",
    strict_mode_config=models.StrictModeConfig(enabled=True, max_query_limit=10),
)

import { QdrantClient } from "@qdrant/js-client-rest";

const client = new QdrantClient({ host: "localhost", port: 6333 });

client.createCollection("{collection_name}", {
  strict_mode_config: {
    enabled: true,
    max_query_limit: 10,
  },
});

use qdrant_client::Qdrant;
use qdrant_client::qdrant::{CreateCollectionBuilder, StrictModeConfigBuilder};

let client = Qdrant::from_url("http://localhost:6334").build()?;

client
    .create_collection(
        CreateCollectionBuilder::new("{collection_name}")
            .strict_mode_config(StrictModeConfigBuilder::default().enabled(true).max_query_limit(10)),
    )
    .await?;

import io.qdrant.client.QdrantClient;
import io.qdrant.client.QdrantGrpcClient;
import io.qdrant.client.grpc.Collections.CreateCollection;
import io.qdrant.client.grpc.Collections.StrictModeConfig;

QdrantClient client =
    new QdrantClient(QdrantGrpcClient.newBuilder("localhost", 6334, false).build());

client
    .createCollectionAsync(
        CreateCollection.newBuilder()
            .setCollectionName("{collection_name}")
            .setStrictModeConfig(
                StrictModeConfig.newBuilder().setEnabled(true).setMaxQueryLimit(10).build())
            .build())
    .get();

using Qdrant.Client;
using Qdrant.Client.Grpc;

var client = new QdrantClient("localhost", 6334);

await client.CreateCollectionAsync(
  collectionName: "{collection_name}",
  strictModeConfig: new StrictModeConfig { Enabled = true, MaxQueryLimit = 10 }
);

import (
  "context"

  "github.com/qdrant/go-client/qdrant"
)

client, err := qdrant.NewClient(&qdrant.Config{
  Host: "localhost",
  Port: 6334,
})

client.CreateCollection(context.Background(), &qdrant.CreateCollection{
  CollectionName: "{collection_name}",
  StrictModeConfig: &qdrant.StrictModeConfig{
    Enabled: qdrant.PtrOf(true),
    MaxQueryLimit: qdrant.PtrOf(uint32(10)),
  },
})

Maximum timeout parameter

Long running operations are often symptomatic of a deeper issue.

Setting max_timeout caps the maximum value in seconds for the timeout parameter in all API operations.

PUT /collections/{collection_name}
{
    "strict_mode_config": {
        "enabled": true,
        "max_timeout": 10
    }
}

curl -X PUT http://localhost:6333/collections/{collection_name} \
  -H 'Content-Type: application/json' \
  --data-raw '{
    "strict_mode_config": {
      "enabled":" true,
      "max_timeout": 10
    }
  }'

from qdrant_client import QdrantClient, models

client = QdrantClient(url="http://localhost:6333")

client.create_collection(
    collection_name="{collection_name}",
    strict_mode_config=models.StrictModeConfig(enabled=True, max_timeout=10),
)

import { QdrantClient } from "@qdrant/js-client-rest";

const client = new QdrantClient({ host: "localhost", port: 6333 });

client.createCollection("{collection_name}", {
  strict_mode_config: {
    enabled: true,
    max_timeout: 10,
  },
});

use qdrant_client::Qdrant;
use qdrant_client::qdrant::{CreateCollectionBuilder, StrictModeConfigBuilder};

let client = Qdrant::from_url("http://localhost:6334").build()?;

client
    .create_collection(
        CreateCollectionBuilder::new("{collection_name}")
            .strict_mode_config(StrictModeConfigBuilder::default().enabled(true).max_timeout(10)),
    )
    .await?;

import io.qdrant.client.QdrantClient;
import io.qdrant.client.QdrantGrpcClient;
import io.qdrant.client.grpc.Collections.CreateCollection;
import io.qdrant.client.grpc.Collections.StrictModeConfig;

QdrantClient client =
    new QdrantClient(QdrantGrpcClient.newBuilder("localhost", 6334, false).build());

client
    .createCollectionAsync(
        CreateCollection.newBuilder()
            .setCollectionName("{collection_name}")
            .setStrictModeConfig(
                StrictModeConfig.newBuilder().setEnabled(true).setMaxTimeout(10).build())
            .build())
    .get();

using Qdrant.Client;
using Qdrant.Client.Grpc;

var client = new QdrantClient("localhost", 6334);

await client.CreateCollectionAsync(
  collectionName: "{collection_name}",
  strictModeConfig: new StrictModeConfig { Enabled = true, MaxTimeout = 10 }
);

import (
  "context"

  "github.com/qdrant/go-client/qdrant"
)

client, err := qdrant.NewClient(&qdrant.Config{
  Host: "localhost",
  Port: 6334,
})

client.CreateCollection(context.Background(), &qdrant.CreateCollection{
  CollectionName: "{collection_name}",
  StrictModeConfig: &qdrant.StrictModeConfig{
    Enabled: qdrant.PtrOf(true),
    MaxTimeout: qdrant.PtrOf(uint32(10)),
  },
})

Maximum size of a filtering condition

Large filtering conditions are expensive to evaluate.

Setting condition_max_size caps the maximum number of element a filtering condition can have.

e.g. the number of elements in MatchAny

PUT /collections/{collection_name}
{
    "strict_mode_config": {
        "enabled": true,
        "condition_max_size": 10
    }
}

curl -X PUT http://localhost:6333/collections/{collection_name} \
  -H 'Content-Type: application/json' \
  --data-raw '{
    "strict_mode_config": {
      "enabled":" true,
      "condition_max_size": 10
    }
  }'

from qdrant_client import QdrantClient, models

client = QdrantClient(url="http://localhost:6333")

client.create_collection(
    collection_name="{collection_name}",
    strict_mode_config=models.StrictModeConfig(enabled=True, condition_max_size=10),
)

import { QdrantClient } from "@qdrant/js-client-rest";

const client = new QdrantClient({ host: "localhost", port: 6333 });

client.createCollection("{collection_name}", {
  strict_mode_config: {
    enabled: true,
    condition_max_size: 10,
  },
});

use qdrant_client::Qdrant;
use qdrant_client::qdrant::{CreateCollectionBuilder, StrictModeConfigBuilder};

let client = Qdrant::from_url("http://localhost:6334").build()?;

client
    .create_collection(
        CreateCollectionBuilder::new("{collection_name}")
            .strict_mode_config(StrictModeConfigBuilder::default().enabled(true).condition_max_size(10)),
    )
    .await?;

import io.qdrant.client.QdrantClient;
import io.qdrant.client.QdrantGrpcClient;
import io.qdrant.client.grpc.Collections.CreateCollection;
import io.qdrant.client.grpc.Collections.StrictModeConfig;

QdrantClient client =
    new QdrantClient(QdrantGrpcClient.newBuilder("localhost", 6334, false).build());

client
    .createCollectionAsync(
        CreateCollection.newBuilder()
            .setCollectionName("{collection_name}")
            .setStrictModeConfig(
                StrictModeConfig.newBuilder().setEnabled(true).setConditionMaxSize(10).build())
            .build())
    .get();

using Qdrant.Client;
using Qdrant.Client.Grpc;

var client = new QdrantClient("localhost", 6334);

await client.CreateCollectionAsync(
  collectionName: "{collection_name}",
  strictModeConfig: new StrictModeConfig { Enabled = true, ConditionMaxSize = 10 }
);

import (
  "context"

  "github.com/qdrant/go-client/qdrant"
)

client, err := qdrant.NewClient(&qdrant.Config{
  Host: "localhost",
  Port: 6334,
})

client.CreateCollection(context.Background(), &qdrant.CreateCollection{
  CollectionName: "{collection_name}",
  StrictModeConfig: &qdrant.StrictModeConfig{
    Enabled: qdrant.PtrOf(true),
    ConditionMaxSize: qdrant.PtrOf(uint64(10)),
  },
})

Maximum number of conditions in a filter

A large number of filtering conditions are expensive to evaluate.

Setting filter_max_conditions caps the maximum number of conditions filters can have.

PUT /collections/{collection_name}
{
    "strict_mode_config": {
        "enabled": true,
        "filter_max_conditions": 10
    }
}

curl -X PUT http://localhost:6333/collections/{collection_name} \
  -H 'Content-Type: application/json' \
  --data-raw '{
    "strict_mode_config": {
      "enabled":" true,
      "filter_max_conditions": 10
    }
  }'

from qdrant_client import QdrantClient, models

client = QdrantClient(url="http://localhost:6333")

client.create_collection(
    collection_name="{collection_name}",
    strict_mode_config=models.StrictModeConfig(enabled=True, filter_max_conditions=10),
)

import { QdrantClient } from "@qdrant/js-client-rest";

const client = new QdrantClient({ host: "localhost", port: 6333 });

client.createCollection("{collection_name}", {
  strict_mode_config: {
    enabled: true,
    filter_max_conditions: 10,
  },
});

use qdrant_client::Qdrant;
use qdrant_client::qdrant::{CreateCollectionBuilder, StrictModeConfigBuilder};

let client = Qdrant::from_url("http://localhost:6334").build()?;

client
    .create_collection(
        CreateCollectionBuilder::new("{collection_name}")
            .strict_mode_config(StrictModeConfigBuilder::default().enabled(true).filter_max_conditions(10)),
    )
    .await?;

import io.qdrant.client.QdrantClient;
import io.qdrant.client.QdrantGrpcClient;
import io.qdrant.client.grpc.Collections.CreateCollection;
import io.qdrant.client.grpc.Collections.StrictModeConfig;

QdrantClient client =
    new QdrantClient(QdrantGrpcClient.newBuilder("localhost", 6334, false).build());

client
    .createCollectionAsync(
        CreateCollection.newBuilder()
            .setCollectionName("{collection_name}")
            .setStrictModeConfig(
                StrictModeConfig.newBuilder().setEnabled(true).setFilterMaxConditions(10).build())
            .build())
    .get();

using Qdrant.Client;
using Qdrant.Client.Grpc;

var client = new QdrantClient("localhost", 6334);

await client.CreateCollectionAsync(
  collectionName: "{collection_name}",
  strictModeConfig: new StrictModeConfig { Enabled = true, FilterMaxConditions = 10 }
);

import (
  "context"

  "github.com/qdrant/go-client/qdrant"
)

client, err := qdrant.NewClient(&qdrant.Config{
  Host: "localhost",
  Port: 6334,
})

client.CreateCollection(context.Background(), &qdrant.CreateCollection{
  CollectionName: "{collection_name}",
  StrictModeConfig: &qdrant.StrictModeConfig{
    Enabled: qdrant.PtrOf(true),
    FilterMaxConditions: qdrant.PtrOf(uint64(10)),
  },
})

Maximum batch size when inserting vectors

Sending very large batch upserts can create internal congestion.

Setting upsert_max_batchsize caps the maximum size in bytes of a batch during vector upserts.

PUT /collections/{collection_name}
{
    "strict_mode_config": {
        "enabled": true,
        "upsert_max_batchsize": 1000
    }
}

curl -X PUT http://localhost:6333/collections/{collection_name} \
  -H 'Content-Type: application/json' \
  --data-raw '{
    "strict_mode_config": {
      "enabled":" true,
      "upsert_max_batchsize": 1000
    }
  }'

from qdrant_client import QdrantClient, models

client = QdrantClient(url="http://localhost:6333")

client.create_collection(
    collection_name="{collection_name}",
    strict_mode_config=models.StrictModeConfig(enabled=True, upsert_max_batchsize=1000),
)

import { QdrantClient } from "@qdrant/js-client-rest";

const client = new QdrantClient({ host: "localhost", port: 6333 });

client.createCollection("{collection_name}", {
  strict_mode_config: {
    enabled: true,
    upsert_max_batchsize: 1000,
  },
});

use qdrant_client::Qdrant;
use qdrant_client::qdrant::{CreateCollectionBuilder, StrictModeConfigBuilder};

let client = Qdrant::from_url("http://localhost:6334").build()?;

client
    .create_collection(
        CreateCollectionBuilder::new("{collection_name}")
            .strict_mode_config(StrictModeConfigBuilder::default().enabled(true).upsert_max_batchsize(1000)),
    )
    .await?;

import io.qdrant.client.QdrantClient;
import io.qdrant.client.QdrantGrpcClient;
import io.qdrant.client.grpc.Collections.CreateCollection;
import io.qdrant.client.grpc.Collections.StrictModeConfig;

QdrantClient client =
    new QdrantClient(QdrantGrpcClient.newBuilder("localhost", 6334, false).build());

client
    .createCollectionAsync(
        CreateCollection.newBuilder()
            .setCollectionName("{collection_name}")
            .setStrictModeConfig(
                StrictModeConfig.newBuilder().setEnabled(true).setUpsertMaxBatchsize(1000).build())
            .build())
    .get();

using Qdrant.Client;
using Qdrant.Client.Grpc;

var client = new QdrantClient("localhost", 6334);

await client.CreateCollectionAsync(
  collectionName: "{collection_name}",
  strictModeConfig: new StrictModeConfig { Enabled = true, UpsertMaxBatchsize = 1000 }
);

import (
  "context"

  "github.com/qdrant/go-client/qdrant"
)

client, err := qdrant.NewClient(&qdrant.Config{
  Host: "localhost",
  Port: 6334,
})

client.CreateCollection(context.Background(), &qdrant.CreateCollection{
  CollectionName: "{collection_name}",
  StrictModeConfig: &qdrant.StrictModeConfig{
    Enabled: qdrant.PtrOf(true),
    UpsertMaxBatchsize: qdrant.PtrOf(uint64(1000)),
  },
})

Maximum collection storage size

It is possible to set the maximum size of a collection in terms of vectors and/or payload storage size.

Setting max_collection_vector_size_bytes and/or max_collection_payload_size_bytes caps the maximum byte size of a collection.

PUT /collections/{collection_name}
{
    "strict_mode_config": {
        "enabled": true,
        "max_collection_vector_size_bytes": 1000000,
        "max_collection_payload_size_bytes": 1000000
    }
}

curl -X PUT http://localhost:6333/collections/{collection_name} \
  -H 'Content-Type: application/json' \
  --data-raw '{
    "strict_mode_config": {
      "enabled":" true,
      "max_collection_vector_size_bytes": 100000,
      "max_collection_payload_size_bytes": 100000
    }
  }'

from qdrant_client import QdrantClient, models

client = QdrantClient(url="http://localhost:6333")

client.create_collection(
    collection_name="{collection_name}",
    strict_mode_config=models.StrictModeConfig(enabled=True, max_collection_vector_size_bytes=1000000, max_collection_payload_size_bytes=1000000),
)

import { QdrantClient } from "@qdrant/js-client-rest";

const client = new QdrantClient({ host: "localhost", port: 6333 });

client.createCollection("{collection_name}", {
  strict_mode_config: {
    enabled: true,
    max_collection_vector_size_bytes: 1000000,
    max_collection_payload_size_bytes: 1000000,
  },
});

use qdrant_client::Qdrant;
use qdrant_client::qdrant::{CreateCollectionBuilder, StrictModeConfigBuilder};

let client = Qdrant::from_url("http://localhost:6334").build()?;

client
    .create_collection(
        CreateCollectionBuilder::new("{collection_name}")
            .strict_mode_config(StrictModeConfigBuilder::default().enabled(true).max_collection_vector_size_bytes(1000000).max_collection_payload_size_bytes(1000000)),
    )
    .await?;

import io.qdrant.client.QdrantClient;
import io.qdrant.client.QdrantGrpcClient;
import io.qdrant.client.grpc.Collections.CreateCollection;
import io.qdrant.client.grpc.Collections.StrictModeConfig;

QdrantClient client =
    new QdrantClient(QdrantGrpcClient.newBuilder("localhost", 6334, false).build());

client
    .createCollectionAsync(
        CreateCollection.newBuilder()
            .setCollectionName("{collection_name}")
            .setStrictModeConfig(
                StrictModeConfig.newBuilder().setEnabled(true).setMaxCollectionVectorSizeBytes(1000000).setMaxCollectionPayloadSizeBytes(1000000).build())
            .build())
    .get();

using Qdrant.Client;
using Qdrant.Client.Grpc;

var client = new QdrantClient("localhost", 6334);

await client.CreateCollectionAsync(
  collectionName: "{collection_name}",
  strictModeConfig: new StrictModeConfig { Enabled = true, MaxCollectionVectorSizeBytes = 1000000, MaxCollectionPayloadSizeBytes = 1000000 }
);

import (
  "context"

  "github.com/qdrant/go-client/qdrant"
)

client, err := qdrant.NewClient(&qdrant.Config{
  Host: "localhost",
  Port: 6334,
})

client.CreateCollection(context.Background(), &qdrant.CreateCollection{
  CollectionName: "{collection_name}",
  StrictModeConfig: &qdrant.StrictModeConfig{
    Enabled: qdrant.PtrOf(true),
    MaxCollectionVectorSizeBytes: qdrant.PtrOf(uint64(1000000)),
    MaxCollectionPayloadSizeBytes: qdrant.PtrOf(uint64(1000000)),
  },
})

Maximum points count

Setting max_points_count caps the maximum number of points for a collection.

PUT /collections/{collection_name}
{
    "strict_mode_config": {
        "enabled": true,
        "max_points_count": 1000
    }
}

curl -X PUT http://localhost:6333/collections/{collection_name} \
  -H 'Content-Type: application/json' \
  --data-raw '{
    "strict_mode_config": {
      "enabled":" true,
      "max_points_count": 1000
    }
  }'

from qdrant_client import QdrantClient, models

client = QdrantClient(url="http://localhost:6333")

client.create_collection(
    collection_name="{collection_name}",
    strict_mode_config=models.StrictModeConfig(enabled=True, max_points_count=1000),
)

import { QdrantClient } from "@qdrant/js-client-rest";

const client = new QdrantClient({ host: "localhost", port: 6333 });

client.createCollection("{collection_name}", {
  strict_mode_config: {
    enabled: true,
    max_points_count: 1000,
  },
});

use qdrant_client::Qdrant;
use qdrant_client::qdrant::{CreateCollectionBuilder, StrictModeConfigBuilder};

let client = Qdrant::from_url("http://localhost:6334").build()?;

client
    .create_collection(
        CreateCollectionBuilder::new("{collection_name}")
            .strict_mode_config(StrictModeConfigBuilder::default().enabled(true).max_points_count(1000)),
    )
    .await?;

import io.qdrant.client.QdrantClient;
import io.qdrant.client.QdrantGrpcClient;
import io.qdrant.client.grpc.Collections.CreateCollection;
import io.qdrant.client.grpc.Collections.StrictModeConfig;

QdrantClient client =
    new QdrantClient(QdrantGrpcClient.newBuilder("localhost", 6334, false).build());

client
    .createCollectionAsync(
        CreateCollection.newBuilder()
            .setCollectionName("{collection_name}")
            .setStrictModeConfig(
                StrictModeConfig.newBuilder().setEnabled(true).setMaxPointsCount(1000).build())
            .build())
    .get();

using Qdrant.Client;
using Qdrant.Client.Grpc;

var client = new QdrantClient("localhost", 6334);

await client.CreateCollectionAsync(
  collectionName: "{collection_name}",
  strictModeConfig: new StrictModeConfig { Enabled = true, MaxPointsCount = 1000 }
);

import (
  "context"

  "github.com/qdrant/go-client/qdrant"
)

client, err := qdrant.NewClient(&qdrant.Config{
  Host: "localhost",
  Port: 6334,
})

client.CreateCollection(context.Background(), &qdrant.CreateCollection{
  CollectionName: "{collection_name}",
  StrictModeConfig: &qdrant.StrictModeConfig{
    Enabled: qdrant.PtrOf(true),
    MaxPointsCount: qdrant.PtrOf(uint64(1000)),
  },
})

Rate limiting

An extremely high rate of incoming requests can have a negative impact on the latency.

Setting read_rate_limit and/or write_rate_limit to cap the maximum number of operations per minute per replica.

When exceeding the maximum number of operations, the client will receive an HTTP 429 error code with a suggested delay before retrying.

PUT /collections/{collection_name}
{
    "strict_mode_config": {
        "enabled": true,
        "read_rate_limit": 1000,
        "write_rate_limit": 100,
    }
}

curl -X PUT http://localhost:6333/collections/{collection_name} \
  -H 'Content-Type: application/json' \
  --data-raw '{
    "strict_mode_config": {
      "enabled":" true,
      "read_rate_limit": 1000,
      "write_rate_limit": 100,
    }
  }'

from qdrant_client import QdrantClient, models

client = QdrantClient(url="http://localhost:6333")

client.create_collection(
    collection_name="{collection_name}",
    strict_mode_config=models.StrictModeConfig(enabled=True, read_rate_limit=1000, write_rate_limit=1000,),
)

import { QdrantClient } from "@qdrant/js-client-rest";

const client = new QdrantClient({ host: "localhost", port: 6333 });

client.createCollection("{collection_name}", {
  strict_mode_config: {
    enabled: true,
    read_rate_limit: 1000,
    write_rate_limit: 100,
  },
});

use qdrant_client::Qdrant;
use qdrant_client::qdrant::{CreateCollectionBuilder, StrictModeConfigBuilder};

let client = Qdrant::from_url("http://localhost:6334").build()?;

client
    .create_collection(
        CreateCollectionBuilder::new("{collection_name}")
            .strict_mode_config(StrictModeConfigBuilder::default().enabled(true).read_rate_limit(1000).write_rate_limit(100)),
    )
    .await?;

import io.qdrant.client.QdrantClient;
import io.qdrant.client.QdrantGrpcClient;
import io.qdrant.client.grpc.Collections.CreateCollection;
import io.qdrant.client.grpc.Collections.StrictModeConfig;

QdrantClient client =
    new QdrantClient(QdrantGrpcClient.newBuilder("localhost", 6334, false).build());

client
    .createCollectionAsync(
        CreateCollection.newBuilder()
            .setCollectionName("{collection_name}")
            .setStrictModeConfig(
                StrictModeConfig.newBuilder().setEnabled(true).setReadRateLimit(1000).setWriteRateLimit(100).build())
            .build())
    .get();

using Qdrant.Client;
using Qdrant.Client.Grpc;

var client = new QdrantClient("localhost", 6334);

await client.CreateCollectionAsync(
  collectionName: "{collection_name}",
  strictModeConfig: new StrictModeConfig { Enabled = true, ReadRateLimit = 1000, WriteRateLimit = 100}
);

import (
  "context"

  "github.com/qdrant/go-client/qdrant"
)

client, err := qdrant.NewClient(&qdrant.Config{
  Host: "localhost",
  Port: 6334,
})

client.CreateCollection(context.Background(), &qdrant.CreateCollection{
  CollectionName: "{collection_name}",
  StrictModeConfig: &qdrant.StrictModeConfig{
    Enabled: qdrant.PtrOf(true),
    ReadRateLimit: qdrant.PtrOf(uint32(1000)),
    WriteRateLimit: qdrant.PtrOf(uint32(100)),
  },
})
Was this page useful?

Thank you for your feedback! 🙏

We are sorry to hear that. 😔 You can edit this page on GitHub, or create a GitHub issue.

On this page: