Make a Video with AI
A video page where you describe what you want in words, pick the model yourself, and can attach a first frame, last frame, or reference images. Work runs in the background, so you can close the page and check later.
สิ่งที่ต้องใส่: text, image · สิ่งที่ได้กลับมา: video
Background jobs, because video takes time
Video takes many times longer than image work, so forcing you to sit on the page makes no sense. Jobs are queued and tracked for you — close the tab, do something else, and come back to the result.
First frame, last frame, and reference images
The first frame is where the clip starts, the last frame where it ends; supply both and the transition is built between them. Reference images only guide style. Attaching images controls the look of the result far more tightly than description alone.
Limits of AI video
You get clips of a few seconds. Hands, fingers, and text in frame still break often. Video is the most expensive kind of everyday AI work, so write a clear instruction the first time and check the estimated price before each run.