List vectors in a vector index using Python SDK V2 - Object Storage Service

Use the ListVectors operation to retrieve all vectors stored in a vector index. This is useful for data auditing, incremental synchronization, and bulk export of vector data. Paging is supported to traverse large datasets.

The Python SDK V2 provides two methods:

list_vectors_paginator() — the SDK handles paging automatically. Use this method in most cases.
list_vectors() — returns one page at a time; you manage paging manually.

Permissions

Alibaba Cloud accounts have full permissions by default. Resource Access Management (RAM) users and RAM roles have no permissions by default. An Alibaba Cloud account owner or administrator must grant the required permissions through a RAM policy or a bucket policy.

API	Action	Description
ListVectors	`oss:ListVectors`	Lists vector data.

Method definitions

list_vectors_paginator() — recommended

list_vectors_paginator(**kwargs) → ListVectorsPaginator

Returns a ListVectorsPaginator object. Call iter_page(request) on it to iterate through all pages — the SDK handles continuation tokens automatically.

For the complete method definition, see list_vectors_paginator.

list_vectors() — manual paging

list_vectors(request: ListVectorsRequest, **kwargs) → ListVectorsResult

Parameter	Type	Description
request	ListVectorsRequest	Request parameters. See ListVectorsRequest.

Returns a ListVectorsResult object containing the current page of vectors. For more information, see list_vectors.

Sample code

Use list_vectors_paginator() to iterate through all vectors without managing pagination tokens.

import argparse
import alibabacloud_oss_v2 as oss
import alibabacloud_oss_v2.vectors as oss_vectors

parser = argparse.ArgumentParser(description="list vectors sample")

parser.add_argument('--region', help='The region in which the bucket is located.', required=True)
parser.add_argument('--endpoint', help='The domain names that other services can use to access OSS')
parser.add_argument('--account_id', help='The account id.', required=True)
parser.add_argument('--bucket', help='The name of the bucket.', required=True)
parser.add_argument('--index_name', help='The name of the vector index.', required=True)

def main():

    args = parser.parse_args()

    # Load credentials from environment variables
    credentials_provider = oss.credentials.EnvironmentVariableCredentialsProvider()

    # Use the SDK's default configuration
    cfg = oss.config.load_default()
    cfg.credentials_provider = credentials_provider
    cfg.region = args.region
    cfg.account_id = args.account_id
    if args.endpoint is not None:
        cfg.endpoint = args.endpoint

    client = oss_vectors.Client(cfg)

    # Create the paginator — the SDK handles continuation tokens automatically
    paginator = client.list_vectors_paginator()

    request = oss_vectors.models.ListVectorsRequest(
        bucket=args.bucket,
        index_name=args.index_name
    )

    # Iterate through all pages; the loop ends when no more vectors remain
    for page in paginator.iter_page(request):
        for vector in page.vectors:
            print(f'Vector: {vector}')

if __name__ == "__main__":
    main()

References

For the complete sample code, see list_vectors.py.