AI image to video with a motion prompt
Animate a still in GenSpicy AI. Use a reference image, describe one main movement, and compare the first short clip before adding detail.

Pick a still with room for the motion
Look at where the subject will go before you upload the image. A figure near the edge has little space to turn or walk, and a tight crop can leave the model guessing what lies beyond it. If the camera will stay fixed, choose a still that already shows the framing you want.
Direct what happens over time
Use the motion prompt for things the still cannot show. Say how the subject moves, what the camera does and how fast the change happens. Give the model a single action to follow on the first run. A breeze and a locked camera are enough to test whether the original composition holds.
Review the first short clip
Watch the beginning and end before adding more movement. If the subject drifts or the crop changes too much, keep the camera still and reduce the action. When you need several beats, generate short shots separately and put them together in an editor.
Read the settings for the selected model
Duration, resolution and support for additional frames vary by model. Check the Peppers estimate before sending the job, and use a source image you have permission to process. Download the clip from your account when it is ready.
A motion prompt to try
The curtain lifts gently in the breeze, then settles. The camera stays fixed and the mug on the windowsill remains still.
Image to video