Video Tasks

Video generation is asynchronous. Create a task, save its ID, poll the status, and download the result after completion.

Task endpoints

The general video task endpoint is:

POST /v1/video/generations

OpenAI-style video requests use:

POST /v1/videos

Both endpoints return a task object rather than the final video file.

Text-to-video

curl https://kaienapi.com/v1/video/generations \\
  -H "Authorization: Bearer <YOUR_API_KEY>" \\
  -H "Content-Type: application/json" \\
  -d '{
    "model": "kling-v2-master",
    "prompt": "A five-second product video with a slow camera push and a clean background",
    "duration": 5,
    "size": "1280x720"
  }'

Image-to-video

curl https://kaienapi.com/v1/videos/generations \\
  -H "Authorization: Bearer <YOUR_API_KEY>" \\
  -H "Content-Type: application/json" \\
  -d '{
    "model": "veo-3.1",
    "prompt": "Animate the scene with a gentle forward camera movement",
    "image": "https://oss.example.com/first-frame.jpg"
  }'

Polling

Poll every few seconds while the task is queued or running. Stop only at a terminal state such as succeeded or failed.

curl "https://kaienapi.com/v1/video/generations/<TASK_ID>" \\
  -H "Authorization: Bearer <YOUR_API_KEY>"

Save the task ID in your application so the user can resume after refreshing the page. Do not create a second task just because one polling request temporarily failed.

Models

Seedance request fields and examples are in Seedance video. Wan 3.0 is in Wan video. Both use POST /v1/videos with fields at the top level of the JSON body. MiniMax H3 768p and 2K are documented in MiniMax H3 768p and 2K. Use image_url for image-to-video, and start_image_url plus end_image_url for first and last frames. Flux 3 Video is documented in Flux 3 Video. Use POST /v1/videos and read url when the task completes.

Downloading results

When the task succeeds, read the returned video URL or file reference and download it to storage you control. Result URLs may be temporary; do not treat them as permanent asset storage.

Getting started

Images and video

Audio

AI tools

API Reference

Task endpoints

Text-to-video

Image-to-video

Polling

Models

Downloading results

​Task endpoints

​Text-to-video

​Image-to-video

​Polling

​Models

​Downloading results

Task endpoints

Text-to-video

Image-to-video

Polling

Models

Downloading results