Image-to-Video

Sora image-to-video API for generating short videos from uploaded images.

---

Overview

• Method: POST
• Path: /v1/videos
• Content-Type: multipart/form-data
• Category: Video Models (Video) / Sora Video

Authentication

• Header: Authorization: Bearer <token>

---

Request Parameters

Field	Type	Required	Description
prompt	string	Yes	Text prompt describing the video content to generate
input_reference	file	Yes	Uploaded reference image; dimensions must exactly match size
model	string	No	Model name, supports sora-2, sora-2-pro; default is sora-2
seconds	string	No	Video length as a string, supports 4, 8, 12; default is 4
size	string	No	Video size, supports 720x1280, 1280x720, 1024x1792, 1792x1024; 1792 sizes only supported by sora-2-pro

---

Request Example

curl Example

curl -X POST "https://api.gpt.ge/v1/videos" \
  -H "Authorization: Bearer $YOUR_TOKEN" \
  -F "prompt=A fierce tiger bursts into a farmhouse yard, the village dog runs away in fear, and a cat jumps out to defeat the tiger." \
  -F "input_reference=@/path/to/image.png" \
  -F "model=sora-2" \
  -F "seconds=4" \
  -F "size=720x1280"

JavaScript (fetch) Example

const formData = new FormData();
formData.append('prompt', 'A fierce tiger bursts into a farmhouse yard, the village dog runs away in fear, and a cat jumps out to defeat the tiger.');
formData.append('input_reference', fileInput.files[0]);
formData.append('model', 'sora-2');
formData.append('seconds', '4');
formData.append('size', '720x1280');

fetch('https://api.gpt.ge/v1/videos', {
  method: 'POST',
  headers: {
    'Authorization': 'Bearer YOUR_TOKEN'
  },
  body: formData
})
  .then(res => res.json())
  .then(console.log);

Python Example (requests)

import requests

with open('/path/to/image.png', 'rb') as f:
    files = {'input_reference': f}
    data = {
        'prompt': 'A fierce tiger bursts into a farmhouse yard, the village dog runs away in fear, and a cat jumps out to defeat the tiger.',
        'model': 'sora-2',
        'seconds': '4',
        'size': '720x1280'
    }

    resp = requests.post(
        'https://api.gpt.ge/v1/videos',
        headers={'Authorization': 'Bearer YOUR_TOKEN'},
        files=files,
        data=data
    )

print(resp.json())

---

Response Fields

Field	Type	Description
id	string	Video task ID
object	string	Returned object type, usually video
created_at	integer	Creation timestamp
status	string	Task status
completed_at	null	Completion time, null if not finished
error	null	Error information, null on success
expires_at	null	Expiration time, currently null
model	string	Model used
progress	integer	Task progress
remixed_from_video_id	null	Remix source video ID, currently null if unused
seconds	string	Video duration
size	string	Video size

---

Success Response Example (200)

{
  "id": "video_68f082321ed08193a4eaf01376fa10bc0284bd663de64dc5",
  "object": "video",
  "created_at": 1760592434,
  "status": "queued",
  "completed_at": null,
  "error": null,
  "expires_at": null,
  "model": "sora-2",
  "progress": 0,
  "remixed_from_video_id": null,
  "seconds": "4",
  "size": "720x1280"
}