Skip to main content

Free & Serverless Clusters

Free and Serverless clusters are serving clusters. Use this page for the basic lifecycle: create, connect, and manage.

📘Note

For Dedicated clusters, see Dedicated Cluster. For on-demand search through a project endpoint, see Connect for On-Demand Search.

Create

Before creating a Free or Serverless cluster, make sure you have registered with Zilliz Cloud and have ownership of the organization or project where the cluster will be created.

📘Note

Each organization can have only one Free cluster. For additional serving clusters, use Serverless or Dedicated.

You can create a Free or Serverless cluster from the Zilliz Cloud console. When the cluster status changes to Running, the cluster is ready. Save the cluster credentials shown during creation because the password is shown only once.

You can also create clusters through the RESTful API.

Create a Free cluster

bash
curl --request POST \
--url "https://api.cloud.zilliz.com/v2/clusters/createFree" \
--header "Authorization: Bearer ${API_KEY}" \
--header "Accept: application/json" \
--header "Content-Type: application/json" \
--data-raw '{
"clusterName": "cluster-free",
"projectId": "proj-xxxxxxxxxxxxxxxxxxxxxx",
"regionId": "gcp-us-west1"
}'

Create a Serverless cluster

bash
curl --request POST \
--url "https://api.cloud.zilliz.com/v2/clusters/createServerless" \
--header "Authorization: Bearer ${API_KEY}" \
--header "Accept: application/json" \
--header "Content-Type: application/json" \
--data-raw '{
"clusterName": "cluster-serverless",
"projectId": "proj-xxxxxxxxxxxxxxxxxxxxxxx",
"regionId": "gcp-us-west1"
}'
ParameterDescription
API_KEYAPI key used to authenticate control-plane API requests.
clusterNameName of the cluster to create.
projectIdID of the project where the cluster will be created.
regionIdID of the cloud region where the cluster will be created.

Connect

Free and Serverless clusters use the following serving endpoint pattern:

bash
https://{cluster-id}.serverless.{region}.vectordb.zillizcloud.com

Copy the cluster public endpoint from the Connect card on the cluster details page. Use either an API key with access to the cluster or a cluster credential in username:password format as the token.

python
from pymilvus import MilvusClient

CLUSTER_ENDPOINT = "YOUR_CLUSTER_ENDPOINT"
TOKEN = "YOUR_CLUSTER_TOKEN"

client = MilvusClient(
uri=CLUSTER_ENDPOINT,
token=TOKEN,
)

To verify the connection, run a lightweight operation such as listing collections.

python
collections = client.list_collections()
print(collections)

Manage

You can manage Free and Serverless clusters from the cluster details page.

OperationFree clusterServerless cluster
RenameSupported.Supported.
ResumeFree clusters are automatically suspended after 7 consecutive days of inactivity and can be resumed at any time.Serverless clusters do not support suspend and resume operations.
Upgrade deployment optionCan be upgraded to Serverless or Dedicated. Free to Dedicated creates a new Dedicated cluster and migrates data from the Free cluster.Can be upgraded to Dedicated. Serverless to Dedicated creates a new Dedicated cluster and migrates data from the Serverless cluster.
DropSupported. Free clusters cannot be restored from the recycle bin after deletion.Supported.

When an upgrade creates a new Dedicated cluster, remember to update the cluster endpoint in your application code.

Drop

To drop a cluster programmatically, call the drop cluster API with the cluster ID.

bash
curl --request POST \
--url "https://api.cloud.zilliz.com/v2/clusters/${CLUSTER_ID}/drop" \
--header "Authorization: Bearer ${API_KEY}" \
--header "Accept: application/json" \
--header "Content-Type: application/json"
Ctrl I