Text to video

You write what should happen in the shot and the model generates it frame by frame. The clearest prompts name the camera move first, then the subject, the action and the setting.

Around 40 to 90 words works best: one shot, one camera move, a clear subject and action. The prompt helper beside the box rewrites a short idea into that shape.

Yes, on models with sound. Put the line in quotation marks and the character speaks it with matching lip movement.

Veo 3.1 for sound and dialogue, Kling 3 Pro for camera work and motion, Seedance for the best price per second. Each model's page compares length, resolution and cost.

Yes. New accounts get 35 free credits with no card, enough for several short clips. After that you can buy a one-off pack (70 credits for $10) or take a plan from $19 a month.