What makes this prompt more controllable?
Runway's current guidance treats the input image as the visual source and the text prompt as motion direction. Begin simply, use positive descriptions and add detail only to fix a visible gap.
Build your runway image-to-video prompt builder
A simple motion test.
The camera slowly arcs to the right as the subject lifts their head and looks toward the light.
How to use this workflow
- Use a clean source image; visual artifacts may become more obvious in motion.
- Generate the Quick prompt first.
- Add one missing motion component at a time: subject, environment, camera, timing or direction.
- For sequences, keep the requested events realistic for the selected duration.
Failure diagnosis and next prompt
Filled editorial example
Illustrative input: a portrait with a visible light beam and floating dust.
The camera slowly arcs to the right as the subject lifts their head and looks toward the light. Dust moves softly through the light beam. The motion begins gently and settles into a steady hold, ending with the subject's face in sharp focus.
Editorial example only — not a Runway output or hands-on test.
Compatibility and limits
Based on Runway's current Gen-4.5 image-to-video guide. Recheck the official guide when the active model changes.
Generation quality and credit use can change by model, version, input asset and settings. Check the product interface and current official documentation before paying for a plan or running a large batch.