Text to video
Describe camera motion, subject action, scene style, duration, and mood.
AI Video Generator
Create AI videos from prompts and images.
Generate concept shots, product motion, short clips, storyboards, and visual experiments with text-to-video and image-to-video models.
- Input
- Prompt or reference image
- Models
- Veo, Kling, Wan, Seedance
- Output
- Short video clips
Use cases
Video creation without leaving the workspace
Write the shot, choose a model, add visual references when needed, and keep the result inside your project.
Image to video
Use generated images, product photos, or concept frames as the starting point.
Model-specific entry points
Jump directly into Veo 3.1, Kling 2.6, Seedance 2, and other video model pages.
Seedance 2 examples
Preview motion styles before generating
Review portrait motion, cinematic landscapes, and interior camera movement to compare pacing, subject stability, and visual style.
Cinematic portrait motion
Useful for judging facial consistency, light behavior, and controlled close-up movement.
Landscape reveal shot
Shows environmental motion, spatial layers, and natural camera drift in a wide scene.
Interior camera move
Useful for evaluating spatial continuity, material detail, and smooth indoor movement.
Workflow
How it works
Write the shot
Define the scene, action, camera movement, and visual treatment.
Choose a model
Choose Veo, Kling, Wan, Seedance, or another available video model.
Generate and reuse
Generate a clip, then use it in projects, workflows, or follow-up generations.
FAQ
Frequently asked questions
Can VAN GOGH generate video from text?
Yes. The video generator can use prompts to drive supported video models.
Can video generation use reference images?
Yes. Image-to-video modes can use visual references to guide the generated result.
How is this different from the Workflow Builder?
This page is a direct video tool entry; the Workflow Builder connects image, video, audio, and other steps into a reusable canvas.