Prompt

Create a realistic video based on the reference photo. Keep both people’s faces, hairstyles, clothing, body proportions, and the original nighttime café setting exactly the same. The man sits with his head slightly lowered, looking at his phone. The woman sits very close to him, gently leans toward him, and quietly talks to him in a natural, intimate way. Girl's movements: subtle and realistic. She slightly moves her lips as if speaking, gently tilts her head toward him, makes a small natural hand gesture, blinks softly, and moves her hair slightly. Occasionally, she looks at his face and then back toward the phone. The man makes only tiny natural movements, occasionally shifting his head slightly while listening. Keep the camera completely stationary, with no zoom, no cuts, and no camera movement. Preserve the warm café lights, reflections in the glass, plants, shadows, and cinematic nighttime atmosphere. Ultra-realistic smartphone video, natural human motion, subtle facial expressions, authentic body language, low-light iPhone camera quality, realistic skin texture. Important: Minimal movement only. No exaggerated gestures, no sudden head movements, no face changes, no body distortion, no extra fingers, no artificial animation, and no changes to the original composition. The scene should feel like a real candid moment recorded on a phone.

Result

Grok Imagine Video 1.5 Preview

Settings

Reference images