> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://docs.nvidia.com/ccluster/llms.txt.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://docs.nvidia.com/ccluster/_mcp/server.

# Python SDK Reference

> Access the NVIDIA CCluster using python

We provide a Python SDK to programmatically access NVIDIA CCluster backend APIs to manage your deployments.

For unattended use in CI or automation, authenticate with a [service account](/ccluster/resources/service-accounts) by setting `CENTML_SERVICE_ACCOUNT_ID` and `CENTML_SERVICE_ACCOUNT_SECRET` before calling the SDK. Interactive development can use `centml login` instead — see [Client Setup](/ccluster/clients/setup).

Following is a sample code to list all the inference deployments under your account.

```python
import centml
from centml.sdk.api import get_centml_client
from centml.sdk import DeploymentType

with get_centml_client() as cclient:
    deployments = cclient.get(DeploymentType.INFERENCE_V3)
    print(deployments)
```

For more examples, please refer to [https://github.com/CentML/centml-python-client/blob/main/examples/sdk/](https://github.com/CentML/centml-python-client/blob/main/examples/sdk/)

## Managing hardware instances

The SDK can also list, create, and delete the **hardware instances** available to a cluster. See [Managing Hardware Instances](/ccluster/resources/hardware-instances) for details.

## What's next

#### [Managing Vault Objects](/ccluster/resources/vault)

Learn how to generate and store NVIDIA CCluster tokens and other vault objects.

#### [Clients](/ccluster/clients/setup)

Learn how to interact with the NVIDIA CCluster programmatically