Image to Video
Bring a still to life.
About Image to Video
Start from a photo, a render or a frame you generated here. The image becomes the first frame, and the prompt says what happens next.
Models you can use
- Lunar Motion 2.5 by Lunar Labs — Our flagship. Thirty-second takes with sound, lip sync and nine-image character lock.
- Veo 3.1 by Google — Photoreal footage with dialogue and sound design, up to native 4K.
- Kling 3.0 Pro by Kuaishou — Fluid human motion and precise start and end frames.
- Seedance 2.0 by ByteDance — Sharp 1080p with references, motion transfer and sound.
- Hailuo H3 by MiniMax — Expressive physics and dramatic action, fast.
- Wan 3.0 by Alibaba — Faithful to long, detailed prompts, with sound.
Direct the camera
Pick from 23 named camera moves — dolly in, crash zoom, orbit, crane, FPV drone — and set the film look: camera body, lens, focal length, aperture, shutter, film stock, lighting and colour palette. See every move.
How it works
- Step 1
Upload a still
A photo, a product shot, a render or an AI image. It becomes the first frame of the clip.
- Step 2
Describe the motion
Say what moves and how — "she turns to the window, hair lifting in the wind" — and add a camera move if you want one.
- Step 3
Generate
The clip keeps the shape of your image and starts exactly on it.
Questions
- Will the video keep my photo exactly?
- The first frame is your image, and the clip follows its shape. What happens after that follows your prompt, so describe what should stay still as well as what should move.
- How much does it cost?
- A 5-second 480p SD clip on Lunar Motion 2.5 is 50 credits; longer, sharper or other models cost more, and the exact price is on the Generate button before you press it. New accounts get 60 free credits to try it.
- Can I use the videos commercially?
- Yes, on any paid plan: Starter, Creator and Studio include commercial use. Free-plan downloads carry a watermark.