Endpoints

Video (async jobs)

Create video generation jobs with POST /videos on api14, poll GET /videos/{id} until they finish and download the finished result.

Try it in the playground

Generate video with Veo 3.1, Sora 2 Pro, Kling, Seedance, Runway, Wan, Hailuo and more. Video generation takes from tens of seconds to several minutes, so it runs as an asynchronous job: you create a job, poll its status, then download the result.

Step	Request
1. Create	`POST /api/v1/videos`
2. Poll	`GET /api/v1/videos/{id}`
3. Download	`GET /api/v1/videos/{id}/content?index=0`

Create a job

POST https://api14.com/api/v1/videos

Parameter	Type	Required	Description
`model`	string	yes	A video model id, e.g. `google/veo-3.1-fast`.
`prompt`	string	yes	What happens in the video.
`duration`	number	no	Length in seconds. Defaults to the model's first supported duration.
`resolution`	string	no	e.g. `720p`, `1080p`. Must be supported by the model.
`aspect_ratio`	string	no	e.g. `16:9`, `9:16`.
`generate_audio`	boolean	no	Generate a soundtrack where supported (default `true`).
`image`	string	no	First frame as a URL or `data:` URI (image-to-video).

Supported durations, resolutions and aspect ratios per model: GET /videos/models (public) or the model page.

Balance check

Before the job is submitted, the gateway estimates its price as duration × the per-second price for the requested resolution (and audio option). If your balance is lower than the estimate the request fails with 402 insufficient_balance and nothing is started.

Response

{
  "id": "vid_01JABCDEF",
  "status": "queued",
  "polling_url": "/api/v1/videos/vid_01JABCDEF"
}

Poll the job

GET https://api14.com/api/v1/videos/{id}

{
  "id": "vid_01JABCDEF",
  "status": "completed",
  "generation_id": "gen-1791100000-abc",
  "unsigned_urls": ["/api/v1/videos/vid_01JABCDEF/content?index=0"],
  "usage": { "cost": 1.2 }
}

`status`	Meaning
`queued`	Waiting for the provider.
`in_progress`	Rendering.
`completed`	Done; `unsigned_urls` lists the outputs.
`failed`, `cancelled`, `expired`	Finished without output; see `error`. You are not charged.

Poll every 5–10 seconds. Only the account that created the job can read it; other ids return 404 not_found.

Download the result

GET https://api14.com/api/v1/videos/{id}/content?index=0

Returns the video bytes (usually video/mp4) with your API key. Output URLs always point to the gateway, never to the upstream provider. If the file is not ready or has expired you get content_unavailable.

Example

import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "https://api14.com/api/v1",
  apiKey: process.env.API14_API_KEY,
});

// Video generation is asynchronous: submit, then poll until completed.
let job = await client.post("/videos", { body: {
  "model": "google/veo-3.1-fast",
  "prompt": "A drone shot over turquoise ocean waves at sunrise",
  "duration": 8,
  "resolution": "720p",
  "aspect_ratio": "16:9"
} });
while (!["completed", "failed"].includes(job.status)) {
  await new Promise((r) => setTimeout(r, 5000));
  job = await client.get(`/videos/${job.id}`);
}
console.log(job.status, job.unsigned_urls);

Billing

You are charged once, when the job reaches completed, using the provider's actual cost. Failed, cancelled and expired jobs are free. Jobs also complete in the background, so a job you never poll is still settled and shows up in Usage.

Errors

Status	Code	When
402	`insufficient_balance`	Balance is below the job's estimated price.
403	`plan_required`	Your plan does not include video models.
404	`model_not_found` / `not_found`	Unknown model, or job id not yours.
502	`upstream_unavailable`	Status could not be read; retry.

See Errors.