GuidesBuild with VideoGenLibraries & SDKs

ChatGPT (custom GPT)

Give a custom GPT the ability to make videos with VideoGen.

Build a custom GPT in ChatGPT that can generate videos, images, voiceovers, sound effects, music, and avatars with VideoGen. The GPT calls the VideoGen API directly through GPT Actions, driven by the hosted OpenAPI document.

Create the custom GPT

  1. In ChatGPT, open Explore GPTs → Create, then the Configure tab.
  2. Under Actions, choose Create new action.
  3. In the schema field, choose Import from URL and paste the OpenAPI document:
https://api.videogen.io/openapi.json
  1. Under Authentication, choose API Key, set Auth Type to Bearer, and paste an API key from app.videogen.io/developers.

That’s it — the GPT can now call every VideoGen endpoint. Give it instructions like: “When asked to make a video, call scriptToVideo, poll getWorkflowRun until it succeeds, then export the project.”

How the API fits GPT Actions

  • Workflows (POST /v1/workflows/*) return a workflowRunId; poll GET /v1/workflows/runs/{workflowRunId} until status is succeeded, failed, or cancelled.
  • Media tools (POST /v1/tools/*) return a toolExecutionId; poll GET /v1/tools/executions/{toolExecutionId}.
  • Export a finished project to MP4 with POST /v1/projects/{projectId}/export.

AI plugin manifest

For plugin hosts that read the legacy manifest, VideoGen serves one at:

https://api.videogen.io/.well-known/ai-plugin.json

It advertises bearer auth and points to the same OpenAPI document above.