Product and campaign shots
Build motion studies and product reveals from a clear creative direction.
AI Video Generator
Create a shot from text, animate a first frame, combine references, or edit existing footage in one focused AI video workspace.
Creative evidence
From a trusted frame to a reusable moving shot
Continue the homepage reference workflow: anchor visual direction, direct motion and camera rhythm, then keep the result ready for the next creation.
01 / WAN 3.0
Reference-frame direction
Combine compatible references to clarify subject, setting, and motion direction.
02 / WAN 3.0
Natural image motion
Start from an approved frame, preserve its composition, and introduce continuous movement.
03 / WAN 3.0
Plan the shot first
Validate camera language, pacing, and atmosphere before committing to production.
Use cases
Create for the formats teams publish every day
Use AI video tools for early visual decisions, repeatable content production, and fast iterations across channels.
Short-form social content
Plan vertical movement, subject action, and camera beats for mobile-first publishing.
Storyboards and previsualization
Turn written scenes and reference frames into moving concepts for review.
Image animation
Extend artwork, portraits, and product stills into controlled moving shots.
Reference-led consistency
Reuse verified assets to keep subjects and visual direction connected across iterations.
Footage transformation
Use the configured video-edit model to direct supported changes to existing clips.
How it works
Create an AI video in three steps
A short path from a clear input to a stored result—without a traditional editing timeline.
- 01
Choose the input mode
Start from text, a first frame, compatible references, or an input video based on the configured model.
- 02
Direct and configure
Describe subject, action, camera, light, and atmosphere; then confirm ratio, resolution, duration, and credits.
- 03
Generate and reuse
Track progress, review the output, download it, or reuse an eligible asset in the next creation.
FAQ
AI video generator FAQ
Which video modes are available?
The page exposes only the Wan modes configured on the server: text to video, image to video, multi reference, and video edit. An unavailable model remains hidden.
Can I use a first and last frame?
Image-to-video accepts one required first frame and, when supported by the configured model, one optional last frame.
How are credits calculated?
Video cost is estimated from duration, resolution, and the configured per-second credit rate. The final estimate appears on the Create button.
What happens if generation fails?
Technical failures use an idempotent refund path so the same job cannot refund twice. The failed state remains visible in Creation History.
Where are finished videos stored?
Completed outputs are copied to the configured durable object storage and linked to your creation record.
Can I reuse a generated result?
Eligible image and video outputs can be selected from Creation History and assigned to a compatible reference role.
Wan 3.0 AI Studio
Direct your next AI video
Choose a real supported mode, add your direction, and create with visible settings and credits.
- Four capability-led modes
- Role-aware references
- Qwen 3.8 Max prompt direction
Prompt
A quiet product reveal with controlled dolly movement and soft studio reflections
Wan video Starting image