# RunPod integration

> GPU cloud platform for AI/ML workloads — manage Pods, Serverless endpoints, templates, and network volumes, and run inference jobs.

- Authentication: apiKey

## Tools (28)

- **Cancel Serverless Job**: Cancel a queued or in-progress Serverless job.
- **Create Container Registry Auth**: Save credentials for a private container registry.
- **Create Network Volume**: Create a persistent network volume in a specific data center.
- **Create Pod**: Provision a new GPU or CPU Pod.
- **Create Serverless Endpoint**: Create a new Serverless endpoint backed by a saved template.
- **Create Template**: Create a reusable Pod or Serverless template (image, env, ports, disks).
- **Delete Container Registry Auth**: Delete a saved container registry credential.
- **Delete Network Volume**: Delete a network volume.
- **Delete Pod**: Permanently terminate and remove a Pod.
- **Delete Serverless Endpoint**: Permanently delete a Serverless endpoint.
- **Delete Template**: Delete a template that is not in use.
- **Get Billing Information**: Retrieve usage and billing for Pods, Serverless endpoints, or network volumes.
- **Get Pod**: Retrieve a single Pod's full configuration and status by ID.
- **Get Serverless Endpoint**: Retrieve a single Serverless endpoint by ID.
- **Get Serverless Endpoint Health**: Get a Serverless endpoint's current worker and job queue stats.
- **List Container Registry Auths**: List saved private container registry credentials.
- **List Network Volumes**: List all persistent network volumes on the account.
- **List Pods**: List all GPU Pods on the account.
- **List Serverless Endpoints**: List all Serverless endpoints on the account.
- **List Templates**: List all Pod and Serverless templates owned by the account.
- **Purge Serverless Endpoint Queue**: Drop all queued (not yet running) jobs on a Serverless endpoint.
- **Run Serverless Job (Async)**: Submit an async inference job to a Serverless endpoint and poll for the result.
- **Run Serverless Job (Sync)**: Submit a Serverless inference job and block until it completes (short jobs only).
- **Start, Stop, Restart, or Reset Pod**: Control a Pod's running state — start, stop, restart, or reset (reimages from container).
- **Update Network Volume**: Rename or grow a network volume.
- **Update Pod**: Modify configurable fields on an existing Pod.
- **Update Serverless Endpoint**: Modify configurable fields on an existing Serverless endpoint.
- **Update Template**: Modify configurable fields on an existing template.

## Related prompts

- [Run your private RunPod model over a Google Sheet nightly](https://www.generalinput.com/prompts/run-your-private-runpod-model-over-a-google-sheet-nightly.md)
- [Auto-stop idle RunPod GPUs and log the savings](https://www.generalinput.com/prompts/auto-stop-idle-runpod-gpus-and-log-the-savings.md)
- [Daily RunPod spend and idle pod digest in Slack](https://www.generalinput.com/prompts/daily-runpod-spend-and-idle-pod-digest-in-slack.md)
- [RunPod serverless health watchdog with Slack alerts](https://www.generalinput.com/prompts/runpod-serverless-health-watchdog-with-slack-alerts.md)

Connect RunPod in General Input: https://www.generalinput.com/apps/runpod