Private Cloud on Qdrant - Vector Database

Setup Private Cloud

info@qdrant.tech (Andrey Vasnetsov) — Mon, 01 Jan 0001 00:00:00 +0000

Qdrant Private Cloud Setup

Requirements

Kubernetes cluster: To install Qdrant Private Cloud, you need a standard compliant Kubernetes cluster. You can run this cluster in any cloud, on-premise or edge environment, with distributions that range from AWS EKS to VMWare vSphere. See Deployment Platforms for more information.
Storage: For storage, you need to set up the Kubernetes cluster with a Container Storage Interface (CSI) driver that provides block storage. For vertical scaling, the CSI driver needs to support volume expansion. For backups and restores, the driver needs to support CSI snapshots and restores.

Permissions: To install the Qdrant Kubernetes Operator you need to have cluster-admin access in your Kubernetes cluster.
Locations: By default, the Qdrant Operator Helm charts and container images are served from registry.cloud.qdrant.io.

Note: You can also mirror these images and charts into your own registry and pull them from there.

Configuration

info@qdrant.tech (Andrey Vasnetsov) — Mon, 01 Jan 0001 00:00:00 +0000

Private Cloud Configuration

The Qdrant Private Cloud helm chart has several configuration options. The following YAML shows all configuration options with their default values:

operator:
 # Amount of replicas for the Qdrant operator (v2)
 replicaCount: 1

 image:
 # Image repository for the qdrant operator
 repository: registry.cloud.qdrant.io/qdrant/operator
 # Image pullPolicy
 pullPolicy: IfNotPresent
 # Overrides the image tag whose default is the chart appVersion.
 tag: ""

 imagePullSecrets:
 - name: qdrant-registry-creds

 nameOverride: ""
 fullnameOverride: "operator"

 # Service account configuration
 serviceAccount:
 create: true
 annotations: {}

 ## Additional labels to add to all resources
 customLabels: {}
 
 # Additional pod annotations
 podAnnotations: {}

 # pod security context
 podSecurityContext:
 runAsNonRoot: true
 runAsUser: 10001
 runAsGroup: 20001
 fsGroup: 30001

 # container security context
 securityContext:
 capabilities:
 drop:
 - ALL
 readOnlyRootFilesystem: true
 runAsNonRoot: true
 runAsUser: 10001
 runAsGroup: 20001
 allowPrivilegeEscalation: false
 seccompProfile:
 type: RuntimeDefault

 # Configuration for the Qdrant operator service to expose metrics
 service:
 enabled: true
 type: ClusterIP
 metricsPort: 9290

 # Configuration for the Qdrant operator service monitor to scrape metrics
 serviceMonitor:
 enabled: false

 # Resource requests and limits for the Qdrant operator
 resources: {}

 # Node selector for the Qdrant operator
 nodeSelector: {}

 # Tolerations for the Qdrant operator
 tolerations: []

 # Affinity configuration for the Qdrant operator
 affinity: {}

 watch:
 # If true, watches only the namespace where the Qdrant operator is deployed, otherwise watches the namespaces in watch.namespaces
 onlyReleaseNamespace: true
 # an empty list watches all namespaces.
 namespaces: []

 limitRBAC: true

 # Configuration for the Qdrant operator (v2)
 settings:
 # Does the operator run inside of a Kubernetes cluster (kubernetes) or outside (local)
 appEnvironment: kubernetes
 # The log level for the operator
 # Available options: DEBUG | INFO | WARN | ERROR
 logLevel: INFO
 # Metrics contains the operator config related the metrics
 metrics:
 # The port used for metrics
 port: 9290
 # Health contains the operator config related the health probe
 healthz:
 # The port used for the health probe
 port: 8285
 # Controller related settings
 controller:
 # The period a forced recync is done by the controller (if watches are missed / nothing happened)
 forceResyncPeriod: 10h
 # QPS indicates the maximum QPS to the master from this client.
 # Default is 200
 qps: 200
 # Maximum burst for throttle.
 # Default is 500.
 burst: 500
 # Features contains the settings for enabling / disabling the individual features of the operator
 features:
 # ClusterManagement contains the settings for qdrant (database) cluster management
 clusterManagement:
 # Whether or not the Qdrant cluster features are enabled.
 # If disabled, all other properties in this struct are disregarded. Otherwise, the individual features will be inspected.
 # Default is true.
 enable: true
 # The StorageClass used to make database and snapshot PVCs.
 # Default is nil, meaning the default storage class of Kubernetes.
 storageClass:
 # The StorageClass used to make database PVCs.
 # Default is nil, meaning the default storage class of Kubernetes.
 #database:
 # The StorageClass used to make snapshot PVCs.
 # Default is nil, meaning the default storage class of Kubernetes.
 #snapshot:
 #volumeAttributesClass:
 # driverName: ebs.csi.aws.com
 # default:
 # name: balanced
 # parameters:
 # iops: 3000
 # throughput: 125
 # template:
 # prefix: qdrant-vac
 # parameters: {}
 # Qdrant config contains settings specific for the database
 qdrant:
 # The config where to find the image for qdrant
 image:
 # The repository where to find the image for qdrant
 # Default is "qdrant/qdrant"
 repository: registry.cloud.qdrant.io/qdrant/qdrant
 # Docker image pull policy
 # Default "IfNotPresent", unless the tag is dev, master or latest. Then "Always"
 #pullPolicy:
 # Docker image pull secret name
 # This secret should be available in the namespace where the cluster is running
 # Default not set
 pullSecretName: qdrant-registry-creds
 # storage contains the settings for the storage of the Qdrant cluster
 storage:
 performance:
 # CPU budget, how many CPUs (threads) to allocate for an optimization job.
 # If 0 - auto selection, keep 1 or more CPUs unallocated depending on CPU size
 # If negative - subtract this number of CPUs from the available CPUs.
 # If positive - use this exact number of CPUs.
 optimizerCpuBudget: 0
 # Enable async scorer which uses io_uring when rescoring.
 # Only supported on Linux, must be enabled in your kernel.
 # See: <https://qdrant.tech/articles/io_uring/#and-what-about-qdrant>
 asyncScorer: false
 # Qdrant DB log level
 # Available options: DEBUG | INFO | WARN | ERROR
 # Default is "INFO"
 logLevel: INFO
 # Default Qdrant security context configuration
 securityContext:
 # Enable default security context
 # Default is false
 enabled: false
 # Default user for qdrant container
 # Default not set
 #user: 1000
 # Default fsGroup for qdrant container
 # Default not set
 #fsUser: 2000
 # Default group for qdrant container
 # Default not set
 #group: 3000
 # Network policies configuration for the Qdrant databases
 networkPolicies:
 # Whether or not NetworkPolicy management is enabled.
 # If set to false, no NetworkPolicies will be created.
 # Default is true.
 enable: true
 ingress:
 - ports:
 - protocol: TCP
 port: 6333
 - protocol: TCP
 port: 6334
 # Allow DNS resolution from qdrant pods at Kubernetes internal DNS server
 egress:
 - ports:
 - protocol: UDP
 port: 53
 # the settings for cloud inference proxy
 inference:
 # inference proxy endpoint
 # when set, the value is passed to Qdrant cluster config as `inference.address` config param
 # if QdrantCluster instance has `.spec.config.inference.enabled` field set
 address: ~
 # Scheduling config contains the settings specific for scheduling
 scheduling:
 # Default topology spread constraints (list from type corev1.TopologySpreadConstraint)
 # Default is an empty list
 topologySpreadConstraints: []
 # Default pod disruption budget (object from type policyv1.PodDisruptionBudgetSpec)
 # Default is not set
 podDisruptionBudget: {}
 # ClusterManager config contains the settings specific for cluster manager
 clusterManager:
 # Whether or not the cluster manager (on operator level).
 # If disabled, all other properties in this struct are disregarded. Otherwise, the individual features will be inspected.
 # Default is false.
 enable: true
 # The endpoint address where the cluster manager can be reached
 endpointAddress: "http://qdrant-cluster-manager"
 # InvocationInterval is the interval between calls (started after the previous call is retured)
 # Default is 10 seconds
 invocationInterval: 10s
 # SyncClustersInterval is the interval between sync-clusters calls (started after the previous call is retured)
 # Default is 10 seconds
 syncClustersInterval: 10s
 # Timeout is the duration a single call to the cluster manager is allowed to take.
 # Default is 30 seconds
 timeout: 30s
 # syncClustersTimeout is the duration a single call to the cluster manager to sync clusters is allowed to take.
 # Default is 10 seconds
 syncClustersTimeout: 10s
 # Specifies overrides for the manage rules
 manageRulesOverrides:
 #dry_run:
 #max_transfers:
 #max_transfers_per_collection:
 #rebalance:
 #replicate:
 # Specifies overrides for the manage rules
 syncClustersRulesOverrides:
 #dry_run:
 #max_downtime_sec:
 #sync_interval_sec:
 # Ingress config contains the settings specific for ingress
 ingress:
 # Whether or not the Ingress feature is enabled.
 # Default is true.
 enable: false
 # Which specific ingress provider should be used
 # Default is KubernetesIngress
 provider: KubernetesIngress
 # The specific settings when the Provider is QdrantCloudTraefik
 qdrantCloudTraefik:
 # Enable tls
 # Default is false
 tls: false
 # Secret with TLS certificate
 # Default is None
 secretName: ""
 # List of Traefik middlewares to apply
 # Default is an empty list
 middlewares: []
 # IP Allowlist Strategy for Traefik
 # Default is None
 ipAllowlistStrategy:
 # Enable body validator plugin and matching ingressroute rules
 # Default is false
 enableBodyValidatorPlugin: false
 # EntryPoints is the list of traefik entry points to use for the ingress route
 # Default is ["web"] or ["websecure"] depending on the TLS setting
 entryPoints: []
 # The specific settings when the Provider is KubernetesIngress
 kubernetesIngress:
 # Name of the ingress class
 # Default is None
 #ingressClassName:
 # TelemetryTimeout is the duration a single call to the cluster telemetry endpoint is allowed to take.
 # Default is 3 seconds
 telemetryTimeout: 3s
 # MaxConcurrentReconciles is the maximum number of concurrent Reconciles which can be run. Defaults to 20.
 maxConcurrentReconciles: 20
 # VolumeExpansionMode specifies the expansion mode, which can be online or offline (e.g. in case of Azure).
 # Available options: Online, Offline
 # Default is Online
 volumeExpansionMode: Online
 # BackupManagementConfig contains the settings for backup management
 backupManagement:
 # Whether or not the backup features are enabled.
 # If disabled, all other properties in this struct are disregarded. Otherwise, the individual features will be inspected.
 # Default is true.
 enable: true
 # Snapshots contains the settings for snapshots as part of backup management.
 snapshots:
 # Whether or not the Snapshot feature is enabled.
 # Default is true.
 enable: true
 # The VolumeSnapshotClass used to make VolumeSnapshots.
 # Default is "csi-snapclass".
 volumeSnapshotClass: "csi-snapclass"
 # The duration a snapshot is retained when the phase becomes Failed or Skipped
 # Default is 72h (3d).
 retainUnsuccessful: 72h
 # MaxConcurrentReconciles is the maximum number of concurrent Reconciles which can be run. Defaults to 1.
 maxConcurrentReconciles: 1
 # ScheduledSnapshots contains the settings for scheduled snapshot as part of backup management.
 scheduledSnapshots:
 # Whether or not the ScheduledSnapshot feature is enabled.
 # Default is true.
 enable: true
 # MaxConcurrentReconciles is the maximum number of concurrent Reconciles which can be run. Defaults to 1.
 maxConcurrentReconciles: 1
 # Restores contains the settings for restoring (a snapshot) as part of backup management.
 restores:
 # Whether or not the Restore feature is enabled.
 # Default is true.
 enable: true
 # MaxConcurrentReconciles is the maximum number of concurrent Reconciles which can be run. Defaults to 1.
 maxConcurrentReconciles: 1

qdrant-cluster-manager:
 replicaCount: 1

 image:
 repository: registry.cloud.qdrant.io/qdrant/cluster-manager
 pullPolicy: IfNotPresent
 # Overrides the image tag whose default is the chart appVersion.
 tag: ""

 imagePullSecrets:
 - name: qdrant-registry-creds
 nameOverride: ""
 fullnameOverride: "qdrant-cluster-manager"

 serviceAccount:
 # Specifies whether a service account should be created
 create: true
 # Automatically mount a ServiceAccount's API credentials?
 automount: true
 # Annotations to add to the service account
 annotations: {}
 # The name of the service account to use.
 # If not set and create is true, a name is generated using the fullname template
 name: ""

 podAnnotations: {}
 podLabels: {}

 podSecurityContext:
 runAsNonRoot: true
 runAsUser: 10001
 runAsGroup: 20001
 fsGroup: 30001

 securityContext:
 capabilities:
 drop:
 - ALL
 readOnlyRootFilesystem: true
 runAsNonRoot: true
 runAsUser: 10001
 runAsGroup: 20001
 allowPrivilegeEscalation: false
 seccompProfile:
 type: RuntimeDefault

 service:
 type: ClusterIP

 networkPolicy:
 create: true

 resources: {}
 # We usually recommend not to specify default resources and to leave this as a conscious
 # choice for the user. This also increases chances charts run on environments with little
 # resources, such as Minikube. If you do want to specify resources, uncomment the following
 # lines, adjust them as necessary, and remove the curly braces after 'resources:'.
 # limits:
 # cpu: 100m
 # memory: 128Mi
 # requests:
 # cpu: 100m
 # memory: 128Mi

 nodeSelector: {}

 tolerations: []

 affinity: {}

qdrant-cluster-exporter:
 image:
 repository: registry.cloud.qdrant.io/qdrant/qdrant-cluster-exporter
 pullPolicy: Always
 # Overrides the image tag. Defaults to the chart appVersion.
 tag: ""

 imagePullSecrets:
 - name: qdrant-registry-creds

 nameOverride: ""
 fullnameOverride: ""

 serviceAccount:
 # Specifies whether a service account should be created
 create: true
 # Annotations to add to the service account
 annotations: {}
 # The name of the service account to use.
 # If not set and create is true, a name is generated using the fullname template
 name: ""

 rbac:
 create: true

 ## Additional labels to add to all resources
 customLabels: {}
 
 podAnnotations: {}

 podSecurityContext:
 runAsNonRoot: true
 runAsUser: 65534
 runAsGroup: 65534
 fsGroup: 65534

 securityContext:
 readOnlyRootFilesystem: true
 runAsNonRoot: true
 runAsUser: 65534
 runAsGroup: 65534

 service:
 enabled: true
 type: ClusterIP
 port: 9090
 portName: metrics

 strategy:
 # Prevents double-scraping by terminating the old pod before creating a new one
 # The pod scrapes a large volume of metrics with high cardinality
 type: Recreate

 resources: {}
 # We usually recommend not setting default resources and to leave this as a conscious
 # choice for the user. This allows charts to run on environments with fewer
 # resources, such as Minikube. If you do want to specify resources, uncomment the following
 # lines, adjust them as necessary, and remove the curly braces after 'resources:'.
 # limits:
 # cpu: 100m
 # memory: 128Mi
 # requests:
 # cpu: 100m
 # memory: 128Mi

 nodeSelector: {}

 tolerations: []

 affinity: {}

 serviceMonitor:
 enabled: false
 honorLabels: true
 scrapeInterval: 60s
 scrapeTimeout: 55s

 # Limit RBAC to the release namespace
 limitRBAC: true

 # Watched Namespaces Configuration
 watch:
 # If true, only the namespace where the exporter is deployed is watched, otherwise it watches the namespaces defined in watch.namespaces
 onlyReleaseNamespace: true
 # an empty list watches all namespaces
 namespaces: []

 # Configuration for the qdrant cluster exporter
 config:
 # The log level for the cluster-exporter
 # Available options: DEBUG | INFO | WARN | ERROR
 logLevel: INFO
 # Controller related settings
 controller:
 # Schedule for the controller to do a forced resync (if watches are missed / nothing happened)
 forceResyncPeriod: 10h
 # Indicates the maximum QPS from this client to the master
 # Default is 200
 qps: 200
 # Maximum burst for throttle.
 # Default is 500.
 burst: 500
 # Maximum number of concurrent reconciliations
 maxConcurrentReconciles: 20
 # Controller's object requeueing interval
 requeueInterval: 30s
 # Exporter Metrics Configuration
 metrics:
 # The port on which the metrics are exposed
 port: 9090
 # The path on which the metrics are exposed
 path: /metrics
 # Exporter Health Check Configuration
 healthz:
 # The port used for the health probe
 port: 8085
 # Qdrant Telemetry and Metrics Cache Configuration
 cache:
 # The period after which the cache is invalidated
 ttl: 60s
 # Qdrant Rest Client Configuration
 qdrant:
 restAPI:
 # The qdrant rest api port
 port: 6333
 # Qdrant API Request Timeout after which requests to Qdrant are canceled if not completed
 timeout: 20s
 # Path where qdrant exposes metrics
 metricsPath: "metrics"
 # Qdrant Telemetry Configuration
 telemetry:
 # Path where qdrant exposes telemetry
 path: "telemetry"
 # The level of details for telemetry
 detailsLevel: 6
 # Whether to anonymize the telemetry data
 anonymize: true

Managing a Cluster

info@qdrant.tech (Andrey Vasnetsov) — Mon, 01 Jan 0001 00:00:00 +0000

Managing a Qdrant Cluster

The most minimal QdrantCluster configuration is:

apiVersion: qdrant.io/v1
kind: QdrantCluster
metadata:
 name: qdrant-a7d8d973-0cc5-42de-8d7b-c29d14d24840
 labels:
 cluster-id: "a7d8d973-0cc5-42de-8d7b-c29d14d24840"
 customer-id: "acme-industries"
spec:
 id: "a7d8d973-0cc5-42de-8d7b-c29d14d24840"
 version: "v1.11.3"
 size: 1
 resources:
 cpu: 100m
 memory: "1Gi"
 storage: "2Gi"

The id should be unique across all Qdrant clusters in the same namespace, the name must follow the above pattern and the cluster-id and customer-id labels are mandatory.

Backups

info@qdrant.tech (Andrey Vasnetsov) — Mon, 01 Jan 0001 00:00:00 +0000

Backups

To create a one-time backup, create a QdrantClusterSnapshot resource:

apiVersion: qdrant.io/v1
kind: QdrantClusterSnapshot
metadata:
 name: "qdrant-a7d8d973-0cc5-42de-8d7b-c29d14d24840-snapshot-timestamp"
 labels:
 cluster-id: "a7d8d973-0cc5-42de-8d7b-c29d14d24840"
 customer-id: "acme-industries" 
spec:
 cluster-id: "a7d8d973-0cc5-42de-8d7b-c29d14d24840"
 retention: 1h

You can also create a recurring backup with the QdrantClusterScheduledSnapshot resource:

apiVersion: qdrant.io/v1
kind: QdrantClusterScheduledSnapshot
metadata:
 name: "qdrant-a7d8d973-0cc5-42de-8d7b-c29d14d24840-snapshot-timestamp"
 labels:
 cluster-id: "a7d8d973-0cc5-42de-8d7b-c29d14d24840"
 customer-id: "acme-industries"
spec:
 scheduleShortId: a7d8d973
 cluster-id: "a7d8d973-0cc5-42de-8d7b-c29d14d24840"
 # every hour
 schedule: "0 * * * *"
 retention: 1h

To resture from a backup, create a QdrantClusterRestore resource:

Logging & Monitoring

info@qdrant.tech (Andrey Vasnetsov) — Mon, 01 Jan 0001 00:00:00 +0000

Configuring Logging & Monitoring in Qdrant Private Cloud

Logging

You can access the logs with kubectl or the Kubernetes log management tool of your choice. For example:

kubectl -n qdrant-private-cloud logs -l app=qdrant,cluster-id=a7d8d973-0cc5-42de-8d7b-c29d14d24840

Configuring log levels: You can configure log levels for the databases individually through the QdrantCluster spec. Example:

apiVersion: qdrant.io/v1
kind: QdrantCluster
metadata:
 name: qdrant-a7d8d973-0cc5-42de-8d7b-c29d14d24840
 labels:
 cluster-id: "a7d8d973-0cc5-42de-8d7b-c29d14d24840"
 customer-id: "acme-industries"
spec:
 id: "a7d8d973-0cc5-42de-8d7b-c29d14d24840"
 version: "v1.11.3"
 size: 1
 resources:
 cpu: 100m
 memory: "1Gi"
 storage: "2Gi"
 config:
 log_level: "DEBUG"

Integrating with a log management system

You can integrate the logs into any log management system that supports Kubernetes. There are no Qdrant specific configurations necessary. Just configure the agents of your system to collect the logs from all Pods in the Qdrant namespace.

API Reference

info@qdrant.tech (Andrey Vasnetsov) — Mon, 01 Jan 0001 00:00:00 +0000

API Reference

auth.qdrant.io/v1alpha1

Package v1alpha1 contains API Schema definitions for the qdrant.io v1alpha1 API group

Resource Types

APIAuthentication

APIAuthentication

APIAuthentication is a configuration for authenticating against Qdrant clusters.

Appears in:

APIAuthenticationList

Field	Description	Default	Validation
`apiVersion` string	`auth.qdrant.io/v1alpha1`
`kind` string	`APIAuthentication`
`metadata` ObjectMeta	Refer to Kubernetes API documentation for fields of `metadata`.
`spec` APIAuthenticationSpec

APIAuthenticationSpec

APIAuthenticationSpec describes the configuration for authenticating against Qdrant clusters.

Changelog

info@qdrant.tech (Andrey Vasnetsov) — Mon, 01 Jan 0001 00:00:00 +0000

Changelog

1.9.4 (2026-01-15)

Component	Version
qdrant-kubernetes-api	v1.22.1
operator	2.9.1
qdrant-cluster-manager	v0.3.15
qdrant-cluster-exporter	1.7.6

Latest validated Qdrant version: 1.16.3

Configurable labels support for cluster-manager
Limit RBAC for cluster-exporter to release namespace

1.9.3 (2025-12-22)

Component	Version
qdrant-kubernetes-api	v1.22.1
operator	2.9.1
qdrant-cluster-manager	v0.3.13
qdrant-cluster-exporter	1.7.6

Latest validated Qdrant version: 1.16.3

Performance and stability improvements
Qdrant 1.16 improvements
Configurable volumeAttributesClassName for storage volumes

1.9.2 (2025-11-18)

Component	Version
qdrant-kubernetes-api	v1.20.0
operator	2.8.3
qdrant-cluster-manager	v0.3.9
qdrant-cluster-exporter	1.7.2

Qdrant 1.16 improvements

1.9.1 (2025-11-14)

Component	Version
qdrant-kubernetes-api	v1.20.0
operator	2.8.1
qdrant-cluster-manager	v0.3.9
qdrant-cluster-exporter	1.7.2

Automated storage class migration
Support for max_payload_index_count in strict mode
Experimental support for read replica clusters
Enable restoring snapshots into another cluster
Performance and stability improvements

1.8.0 (2025-08-08)

Component	Version
qdrant-kubernetes-api	v1.17.2
operator	2.6.8
qdrant-cluster-manager	v0.3.7
qdrant-cluster-exporter	1.7.1

Add qdrant-cluster-exporter to provide OpenTelemetry/Prometheus metrics for Qdrant clusters

1.7.3 (2025-08-04)

Component	Version
qdrant-kubernetes-api	v1.17.2
operator	2.6.8
qdrant-cluster-manager	v0.3.7

Configurable max_request_size_mb

1.7.2 (2025-06-27)

Component	Version
qdrant-kubernetes-api	v1.17.0
operator	2.6.4
qdrant-cluster-manager	v0.3.7

Performance and stability improvements
Configurable maxCollection

1.7.1 (2025-06-03)

Component	Version
qdrant-kubernetes-api	v1.16.6
operator	2.6.0
qdrant-cluster-manager	v0.3.6

Performance and stability improvements

1.7.0 (2025-05-14)

Component	Version
qdrant-kubernetes-api	v1.16.3
operator	2.4.2
qdrant-cluster-manager	v0.3.5

Add optional automatic shard balancing
Set strict mode by default for new clusters to only allow queries with payload filters on fields that are indexed

1.6.4 (2025-04-17)

Component	Version
qdrant-kubernetes-api	v1.15.5
operator	2.3.4
qdrant-cluster-manager	v0.3.4

Fix bug in operator Helm chart that caused role binding generation to fail when using watch.namespaces

1.6.3 (2025-03-28)

Component	Version
qdrant-kubernetes-api	v1.15.0
operator	2.3.3
qdrant-cluster-manager	v0.3.4

Performance and stability improvements for collection re-sharding

1.6.2 (2025-03-21)

Component	Version
qdrant-kubernetes-api	v1.15.0
operator	2.3.2
qdrant-cluster-manager	v0.3.3

Allow disabling NetworkPolicy management in Qdrant Cluster operator

1.6.1 (2025-03-14)

Component	Version
qdrant-kubernetes-api	v1.14.2
operator	2.3.2
qdrant-cluster-manager	v0.3.3

Add support for GPU instances
Experimental support for automatic shard balancing

1.5.1 (2025-03-04)

Component	Version
qdrant-kubernetes-api	v1.12.0
operator	2.1.26
qdrant-cluster-manager	v0.3.2

Fix scaling down clusters that have TLS with self-signed certificates configured
Various performance improvements and stability fixes

1.5.0 (2025-02-21)

Component	Version
qdrant-kubernetes-api	v1.12.0
operator	2.1.26
qdrant-cluster-manager	v0.3.0

Added support for P2P TLS configuration
Faster node removal on scale down
Various performance improvements and stability fixes

1.4.0 (2025-01-23)

Component	Version
qdrant-kubernetes-api	v1.8.0
operator	2.1.26
qdrant-cluster-manager	v0.3.0

Support deleting peers on horizontal scale down, even if they are already offline
Support removing partially deleted peers

1.3.0 (2025-01-17)

Component	Version
qdrant-kubernetes-api	v1.8.0
operator	2.1.21
qdrant-cluster-manager	v0.2.10

Support for re-sharding with Qdrant >= 1.13.0

1.2.0 (2025-01-16)

Component	Version
qdrant-kubernetes-api	v1.8.0
operator	2.1.20
qdrant-cluster-manager	v0.2.9

Performance and stability improvements

1.1.0 (2024-12-03)

Component	Version
qdrant-kubernetes-api	v1.6.4
operator	2.1.10
qdrant-cluster-manager	v0.2.6

Activate cluster-manager for automatic shard replication

1.0.0 (2024-11-11)

Component	Version
qdrant-kubernetes-api	v1.2.7
operator	0.1.3
qdrant-cluster-manager	v0.2.4

Initial release