Open Media.io Image to Video and upload a well-lit full-body reference photo. Describe a solo dancer snapping their fingers, a second character entering from the right, and both dancing together. If your model supports multiple references, add a friend or pet photo; otherwise prepare a combined scene image. Generate and review identity consistency, movement, and timing. Results depend on the selected model.

Try: “Vertical 9:16 full-body phone video in a bright living room, fixed camera, one adult dancer on the left. The dancer performs small rhythmic steps and arm rolls, then snaps their fingers. On the beat, a second adult dancer enters from the right as if by magic. Both perform simple synchronized shoulder bounces, wrist flicks, and side steps. Keep both faces consistent, two people only, natural hands, stable background, no captions or logos.” For a twin, specify the same face with contrasting outfits; for a pet, describe the animal instead of a second person. Complex choreography may need shorter clips.

Yes, these are creative variations you can describe with Media.io prompts. For a twin dance, start from one clear portrait and request a matching lookalike who joins after the snap. For an Abra Kadabra pet AI video, use a sharp whole-body pet image and request playful animated movements while preserving species, fur, and paws. Model support for multiple reference photos varies, and exact synchronization is not guaranteed.

The duo dance is associated with a Brazilian funk audio hook repeating “abra kadabra,” not Lady Gaga’s 2025 song “Abracadabra.” You can generate a visual dance scene with Media.io and add suitable audio in a video editor or on the posting platform. Music rights and platform rules still apply; an AI-generated clip does not automatically include permission to reuse the original trending sound.