Ramp subject replacement
One photo replaces the moving hero, while the built-in ramp supplies the whole stunt.
Flying Car AI: Turn One Photo into a Ramp Stunt
Upload one complete subject photo for a Flying Car AI ramp video. Keep the reference scene and motion, watch the panda-on-horse result and review credits.
Flying Car AI replaces the moving hero in a built-in night-arena stunt. The ramp, spectators, background cars, camera sequence and action order remain the guide. Your photo supplies the new subject: a car, an animal, or a complete rider-and-mount combination. Flying Roadster AI is one vehicle use case; the panda riding a white horse shows a different subject using the same ramp.
Try this effect →Create your video
One photo replaces the moving hero, while the built-in ramp supplies the whole stunt.
Checking your sign-in…
JPG, PNG or WebP, up to 10 MB. Use one clear subject per photo.
Built-in motion reference · rounded duration used for billing · 15 seconds
Wan 3.0 · 480P · The quote includes the built-in reference and 15-second output. Review the displayed credits before generating.
15-second silent video. The template supplies the motion; your photo supplies the complete hero identity.
Read the English motion prompt
Edit Video1. Replace ONLY the single moving hero van with the complete subject or complete rider-and-mount assembly in Image1. Image1 supplies identity, body shape, clothing or fur, accessories and the entire mount when present; never copy its background. Preserve the night stunt arena, tall red-edged curved ramp, parked background cars, spectators, lighting, all camera movements, shot sequence and original action timing. Track the same hero through the ground approach, full ramp climb, airborne stunt, upright smoke landing and final showcase. For a rider and horse, preserve their connection throughout and use coherent natural legs; never add vehicle parts or wheels to an animal. Replace no background subject. No new location, extra action, duplicate hero, cuts to the photo, text or soundtrack. Complete the original full stunt and ending within 15 seconds.
Extra direction and the sound preference are added at submission. Model guidance does not guarantee frame-by-frame matching.
Estimated credits：23 credits
Sign in to createFlying Car AI: Turn One Photo into a Ramp Stunt · Prepare the inputs
Prepare one JPG, PNG or WebP up to 10 MB. Show the whole subject with clear edges and space around it. Include all wheels for a car, all hooves and the tail for a horse, or both the rider and the entire mount in a single photo. A headshot cannot define a full rider-and-horse assembly. Use a photo you own or may edit.
Choose a variation
How to make the video
1. Sign in and upload one complete subject photo. 2. Keep the built-in ramp template; no prompt or reference upload is needed. 3. Review the displayed credit quote and generate. 4. Follow the saved task, then play and download the completed result. The workflow uses Wan 3.0, native 480P, a 15-second silent output and 4:3 framing. A result can vary from the example.
Review the result
Watch the whole clip: approach, ramp climb, airborne stunt, upright smoke landing and final presentation. Track the same face or markings throughout. For a rider and mount, check the saddle connection, all legs and reins after the turn. Check that background spectators and parked cars have not become extra copies of your subject.
If limbs merge or the rider separates, choose a clearer whole-assembly photo with less overlap. If the photo background follows the subject, use a simpler input. Avoid cropped hooves, hidden wheels and several competing subjects. Keep your existing task when checking progress; starting a new task is a separate generation.
Common questions
Can I use something other than a sports car?
Yes. Use one complete subject, including an animal or rider with a mount. Keep a rider and horse together in one reference photo. The example shows a panda riding a white horse; exact identity and movement are not guaranteed.
Do I need to write a prompt or upload the stunt video?
No. The template supplies the ramp reference and instructions. You upload one subject photo, review the credit quote and start generation.
Is the example my generated video, and how is the cost calculated?
The panda-on-horse clip is an existing result video, not your personal output. Your result appears only after your task succeeds. The quote includes the reference-video time and the 15-second output, using the site credit policy; browsing the sample costs no credits.
The generated motion, likeness and soundtrack need to be reviewed before downloading.
Read this page as Markdown