Implementation:Openai Openai node Videos Resource

Knowledge Sources	Openai_Openai_node
Domains	SDK, Video Generation
Last Updated	2026-02-15 12:00 GMT

Overview

The Videos class provides API resource methods for creating, retrieving, listing, deleting, downloading, and remixing AI-generated videos using models like Sora.

Description

The Videos class extends APIResource and exposes six methods covering the full video generation lifecycle. The create method initiates a video generation job with a text prompt, optional input reference image, model selection (defaulting to sora-2), clip duration (4, 8, or 12 seconds), and output resolution. The retrieve method fetches a video job's current status and metadata. The list method returns cursor-paginated video jobs with optional sort ordering.

The delete method removes a video and returns a VideoDeleteResponse confirmation. The downloadContent method retrieves the actual video binary content, with a variant parameter to select between the video MP4, thumbnail, or spritesheet. The remix method creates a new video based on an existing one with an updated prompt.

The Video interface describes the video job object with fields for status tracking (status: queued/in_progress/completed/failed, progress: 0-100), metadata (model, prompt, seconds, size), timestamps (created_at, completed_at, expires_at), error information, and remix provenance (remixed_from_video_id). Supported models include sora-2 and sora-2-pro variants.

Usage

Use this resource to generate videos from text prompts, monitor generation progress, download completed videos, or create remixed variations of existing videos. It is the primary interface for interacting with OpenAI's video generation capabilities.

Code Reference

Source Location

• Repository: openai-node
• File: src/resources/videos.ts

Signature

export class Videos extends APIResource {
  create(body: VideoCreateParams, options?: RequestOptions): APIPromise<Video>;
  retrieve(videoID: string, options?: RequestOptions): APIPromise<Video>;
  list(query?: VideoListParams | null, options?: RequestOptions): PagePromise<VideosPage, Video>;
  delete(videoID: string, options?: RequestOptions): APIPromise<VideoDeleteResponse>;
  downloadContent(videoID: string, query?: VideoDownloadContentParams | null, options?: RequestOptions): APIPromise<Response>;
  remix(videoID: string, body: VideoRemixParams, options?: RequestOptions): APIPromise<Video>;
}

export interface Video {
  id: string;
  completed_at: number | null;
  created_at: number;
  error: VideoCreateError | null;
  expires_at: number | null;
  model: VideoModel;
  object: 'video';
  progress: number;
  prompt: string | null;
  remixed_from_video_id: string | null;
  seconds: VideoSeconds;
  size: VideoSize;
  status: 'queued' | 'in_progress' | 'completed' | 'failed';
}

export type VideoModel = string | 'sora-2' | 'sora-2-pro' | ...;
export type VideoSeconds = '4' | '8' | '12';
export type VideoSize = '720x1280' | '1280x720' | '1024x1792' | '1792x1024';

Import

import OpenAI from 'openai';

I/O Contract

Inputs

Name	Type	Required	Description
body.prompt	string	Yes (create/remix)	Text prompt describing the video to generate.
body.model	VideoModel	No (create)	Video generation model; defaults to sora-2.
body.seconds	'8' | '12'	No (create)	Clip duration in seconds; defaults to 4.
body.size	VideoSize	No (create)	Output resolution (e.g., '1280x720'); defaults to '720x1280'.
body.input_reference	Uploadable	No (create)	Optional image reference to guide generation.
videoID	string	Yes (retrieve/delete/download/remix)	The video job identifier.
query.variant	'thumbnail' | 'spritesheet'	No (downloadContent)	Which downloadable asset to return; defaults to video MP4.
query.order	'desc'	No (list)	Sort order by timestamp.
options	RequestOptions	No	Additional request configuration.

Outputs

Name	Type	Description
id	string	Unique video job identifier.
status	'in_progress' | 'completed' | 'failed'	Current lifecycle status.
progress	number	Approximate completion percentage (0-100).
model	VideoModel	The model used for generation.
prompt	null	The text prompt used.
seconds	VideoSeconds	Clip duration.
size	VideoSize	Output resolution.
created_at	number	Unix timestamp when the job was created.
completed_at	null	Unix timestamp when the job finished.
expires_at	null	Unix timestamp when downloadable assets expire.
error	null	Error details if generation failed.
remixed_from_video_id	null	Source video ID if this is a remix.

Usage Examples

import OpenAI from 'openai';

const client = new OpenAI();

// Create a video
const video = await client.videos.create({
  prompt: 'A serene mountain lake at sunrise with mist rising from the water',
  model: 'sora-2',
  seconds: '8',
  size: '1280x720',
});
console.log(video.id, video.status);

// Poll for completion
let result = await client.videos.retrieve(video.id);
while (result.status !== 'completed' && result.status !== 'failed') {
  await new Promise((r) => setTimeout(r, 5000));
  result = await client.videos.retrieve(video.id);
  console.log(`Progress: ${result.progress}%`);
}

// Download the completed video
const response = await client.videos.downloadContent(video.id, { variant: 'video' });

// List all videos
for await (const v of client.videos.list({ order: 'desc' })) {
  console.log(v.id, v.status);
}

// Create a remix
const remix = await client.videos.remix(video.id, {
  prompt: 'Same scene but during a thunderstorm with dramatic lighting',
});

// Delete a video
const deleted = await client.videos.delete(video.id);
console.log(deleted.deleted); // true

Related Pages

• Environment:Openai_Openai_node_Node_20_Runtime