Flying Car Text to Video

It is good at a single, legible action in a strong environment: a ramp launch, a low pass over traffic, a climb between buildings, a landing on a rooftop. It is also good at cinematic light and at locations that would be expensive to film. Those are the shots worth generating first.

It is weaker at precise choreography, readable text, number plates and exact real-world geography. If the clip must show a particular car or a particular street, use image-to-video with a photo instead. Text-to-video is for the idea, not the identity.

Write the subject, the action, the camera and the light. For example: “a silver flying car climbs a curved ramp, launches, rotates once and lands on all four wheels; wide side tracking shot; late-afternoon sun; dust and a soft shadow.” That is enough for a coherent five-second beat.

Add physics words. Flying cars have no real-world prior, so the model needs cues: thrusters glowing, landing gear folding, suspension compressing, dust ring, heat haze. Without those cues the car tends to float like a cut-out. Add one camera instruction and one atmosphere instruction, not five.

Generate a short draft at the lowest resolution that lets you judge composition. If the camera is wrong, change the camera sentence. If the car is wrong, change the subject sentence. Do not raise the resolution until the motion reads.

A standard 12-second 720p clip costs 30 credits. A 1080p version costs more because the upstream render is heavier. The pricing page lists the exact packs, and the studio shows the cost before you press Generate.