What it does
- Text to video generates cinematic action and environments from a detailed natural-language prompt.
- Image to video adds controlled subject and camera motion to an existing key visual.
- Complex motion generation lets you describe choreography, interacting subjects, speed, weight, and environmental response.
- Cinematic camera control supports tracking, orbit, crane, close-up, wide, and reveal shots.
- Physical detail and materials render weight, texture, lighting, weather, and environmental interaction.
- Controlled image animation carries the composition and character identity of a source image into motion.
Pricing
The free plan includes 10 daily credits. Credit packs are available for purchase.
Good to know
- Runs in a web browser.
- Output aspect ratios include landscape, portrait, square, and cinematic.
Questions & answers
- What is Wan 3 AI?
- Wan 3 is a video-generation workflow focused on text-to-video, image-to-video, complex movement, and cinematic visual control.
- Can I create from a still image?
- Yes. Upload reference images and describe the desired camera, subject, and environmental motion.
- Which aspect ratios are available?
- Choose landscape, portrait, square, or cinematic output depending on the publishing destination.
- How are failed tasks handled?
- Reserved credits are refunded once for eligible provider failures.
In their words
Create cinematic video from text and images with controlled movement, convincing materials, and clear spatial composition.
Direct the physics of the shot.