Create 5- or 10-second AI video clips from a prompt and an optional reference image. Choose the frame, resolution, and duration, then refine the result online.
AI Video Generator
Describe the shot, the movement, and the mood. Add a reference image when you want the first frame to follow a specific visual direction.
The generator keeps the workflow focused on the source, the motion, and the final frame rather than a long editing timeline.
Build a Short Video Around One Clear Idea
Start with words, add an image when composition matters, and describe how the scene should move.
- 1
1. Frame the idea
Begin with a text prompt, or upload a JPEG, PNG, or WebP image up to 24MB when you want the clip to follow a particular subject or composition.
- 2
2. Direct the movement
Describe what changes over time: the subject action, camera movement, pace, lighting, atmosphere, and the moment the shot should emphasize.
- 3
3. Choose the format
Select 16:9, 9:16, or 1:1 and set a duration of 5 or 10 seconds. Videos are generated at 720p.
- 4
4. Review and refine
Generate the clip, watch how the motion develops, and revise one part of the prompt at a time when you want a more precise result.
Creative Control at the Decisions That Matter
The generator keeps the workflow focused on the source, the motion, and the final frame rather than a long editing timeline.
Text-First or Image-Guided
Use only a prompt for an open-ended scene, or add a reference image when the subject, layout, or visual identity should anchor the clip.
Motion Written Into the Prompt
Direct subject movement, camera behavior, pace, and atmosphere in the same place, so the generator has a clear sequence to interpret.
Formats for Different Frames
Choose landscape, portrait, or square output for drafts, feeds, presentations, and short-form stories.
A Workflow Made for Iteration
Keep recent generations together, compare how prompt changes affect motion, and return to promising directions without losing the earlier result.
AI Video Generator Questions
What to provide, which settings are available, and how to get a more deliberate short clip.
Do I need an image to create a video?
No. A text prompt is enough to begin. Add a reference image only when you want to guide the subject, composition, or overall visual direction.
Which reference image files can I upload?
The uploader accepts JPEG, JPG, PNG, and WebP images up to 24MB. Use a clear image that already contains the subject and framing you want to carry forward.
What should a useful video prompt include?
Name the subject, action, camera movement, pace, setting, lighting, and mood. Focus on one short moment instead of trying to fit an entire story into one clip.
Which formats can I generate?
Choose a 16:9, 9:16, or 1:1 aspect ratio and a duration of 5 or 10 seconds. Output resolution is fixed at 720p.
How does a reference image affect the result?
It gives the generator a visual anchor for the opening composition and subject. The prompt should then explain what moves, what stays stable, and how the camera behaves.
Why can repeated generations look different?
Each generation interprets the prompt again. Even with the same source image, motion, timing, framing, and small visual details can vary.
How can I improve a result that feels unfocused?
Shorten the prompt to one main action, make the camera direction explicit, remove conflicting style cues, and change only one instruction before generating again.
Where can I find videos I generated earlier?
Recent Videos shows the latest tasks on this page. Open your asset history when you want to revisit completed clips from earlier sessions.
Turn One Visual Idea Into a Moving Shot
Write the moment, guide the frame if needed, and see how the scene develops over five or ten seconds.