OpenAI Videos Create
openai-videos-createRead-only
Create a video from a text prompt with optional image reference via OpenAI's API. Returns the video job object and downloads video, thumbnail, or spritesheet.
Instructions
Create a video generation job using the OpenAI Videos API. Returns structuredContent with the OpenAI Video job object, and (optionally) downloaded assets as MCP content blocks (tool_result=resource_link|resource).
Input Schema
TableJSON Schema
| Name | Required | Description | Default |
|---|---|---|---|
| prompt | Yes | Text prompt that describes the video to generate (max 32K chars). | |
| input_reference | No | Optional image reference: HTTP(S) URL, base64 / data URL, or file path. | |
| input_reference_fit | No | How to fit input_reference to the requested video size: match (require exact), cover (crop), contain (pad), stretch (distort). Default: contain. | contain |
| input_reference_background | No | Padding background when input_reference_fit=contain: blur (default), black, white, or hex color (#RRGGBB/#RRGGBBAA). | blur |
| model | No | Video model to use (default: sora-2-pro). | sora-2-pro |
| seconds | No | Clip duration in seconds (default: 4 seconds). | |
| size | No | Output resolution formatted as width x height. Sizes 1024x1792 and 1792x1024 require sora-2-pro. | |
| wait_for_completion | No | If true, poll the job until completed/failed (then optionally download assets). Default: true. | |
| timeout_ms | No | Max time to wait for completion when wait_for_completion is true (default: 900000). | |
| poll_interval_ms | No | Polling interval when wait_for_completion is true (default: 2000). | |
| download_variants | No | Which downloadable assets to fetch when completed (default: ['video']). | |
| tool_result | No | Controls content[] shape: 'resource_link' (default) emits ResourceLink items, 'resource' emits EmbeddedResource blocks with base64 blob. | resource_link |