Choose a model
Choose a generation model suited for video creation.
Add references
Upload images as visual references for video generation.
Describe motion
Describe subject motion, camera movement, and scene changes.
Set output
Set the aspect ratio, resolution, and duration, then generate the video.
1 / 4
Beginner's Guide
Use ready-made templates, explore community creations, or get inspired by our asset library
Cinematic product launchStarting from a static product photo of [img1]. Begin in an extreme close-up of the most distinctive texture or detail of the product, then slowly pull back in a smooth cinematic dolly-out, ending with the full product centered and visible in frame. The camera movement is smooth and controlled, with no shake or drift. Native audio: clean and minimal. A single subtle resonance of the product material - glass, leather, metal or fabric - faint and precise, no organic sounds, no moisture, no food texture. Pure silent room tone underneath. No foley layers, no ambient environment, no voiceover, no music. Soft diffused natural light. Premium product aesthetic.
Multilingual spokesperson ad[img2] stands in front of a soft white background, holding [img1] in their hand at chest height, facing the camera directly. The character speaks naturally to the camera in (TYPE A LANGUAGE): "A cold-pressed blend of pomegranate, beetroot, ginger and lemon. No sugar, no preservatives, no heat treatment. Supports liver function and reduces inflammation - naturally. One bottle, every morning." Lips perfectly synced with the audio. Warm diffused natural lighting. Static camera, no shake, no drift.
Storyboard to videoMatch the reference video exactly (timing, moves and audio) while keeping the new character unchanged in appearance, outfit, and identity.
UGC product videoUGC-style selfie video: the person holds and presents the product naturally to camera, speaking with accurate lip-sync in a friendly tone (e.g., "Hiii! Today I'm going to show you this new product that just arrived"), with believable gestures, subtle handheld motion, and quick product close-ups while keeping the person and product unchanged.