Reference image workflow

Wan 3.0 Image to Video Prompt Generator

Upload a reference frame and build an image-to-video prompt specifically for Wan 3.0, with identity locks, motion planning, camera direction and continuity safeguards.

Reference preview
Describe the desired movement, transformation or story beat. Any language is fine.
Best for product shots, characters and consistent image-to-video animation.
Lock these reference elements
I

Your image-to-video plan will appear here

GenVideoKit will analyze the visible starting state, detect instruction/reference conflicts, isolate what should stay consistent and build three model-ready prompt variants.

Reference / instruction conflict
The text request conflicts with the uploaded image.

Reference analysis
AI vision

Image motion plan

Visible starting state
Identity / subject lock
Composition / environment
Motion plan
Visual style
Camera plan
Locked elements applied
Model-ready prompt variants

Prompt

Avoid

Negative / stability instructions

Production workflow

Continue this reference into a full project

Expand this shot into a multi-scene storyboard or estimate the cost of a single-shot project.

This tool prepares prompts only; it does not render video. Your reference image is sent to the configured vision model during the request and is not saved by this tool.
Reference workflow

Wan 3.0 image-to-video prompting

Current Wan 3.0 general video API.

Model-specific direction: Describe composition, subject motion, environment motion, camera behavior and temporal continuity explicitly. Keep cause-and-effect clear, maintain stable identity/geometry, and specify reference relationships when used. Audio direction may be included for current multimodal Wan workflows.

Reference controls

This landing uses the same vision workflow as the general tool, but starts with Wan 3.0 selected.

Text → video Image → video Video → video Native audio References First / last frame
Text prompt generator Cost calculator Model guide