RunningHub API icon

RunningHub API

Interact with the RunningHub API

AI App → Run AI App

AI-generated

Summary

Starts a RunningHub AI application task by POSTing the app ID and node info list to /task/openapi/ai-app/run, returning RunningHub's raw JSON response for the submitted task.

Inputs

  • Webapp ID (required) — ID of the RunningHub AI application to run.
  • Node Info List (required) — JSON array of node information (the app's input overrides). It is parsed with jsonParse and sent as nodeInfoList. Invalid JSON causes an error.
  • Webhook URL — Callback URL for task completion. Sent only if non-empty.
  • Instance Type — Runtime instance: Default (24G VRAM) or Plus (48G VRAM). Defaults to 'default' and is sent when set.
  • Access Password — Password for AI apps that have access encryption enabled. Sent only if non-empty.

Output shape

One JSON item per input item, containing the RunningHub API response for the submitted task. The source does not transform this response for this operation.

The call is asynchronous submission only. It doesn't wait for results, so use the Task Query operations (status or output) or a webhook to get the final output. The API key is sent in the request body and as a Bearer header. A valid license key in the credentials is checked before running. With Continue On Fail, errors are returned on the item instead of stopping the workflow.

Examples

Example 1: Run an image-generation AI app with a custom prompt and get notified on completion

Webapp ID = app ID; Node Info List = [{"nodeId":"6","fieldName":"text","fieldValue":"a cat"}]; Webhook URL = your n8n webhook endpoint

Example 2: Run a heavy app on the larger GPU for a password-protected app

Webapp ID set; Node Info List set; Instance Type = Plus (48G VRAM); Access Password = app password

Links

Discussion