Choose the right operation
Start with a subject, action, setting and one camera movement. Text-to-video creates a new shot; it does not require an existing photograph or video.
How it works
1. Choose the model and input
Choose an available model. Add its required image or video before requesting a price. Only supported settings are shown.
2. Describe the shot or edit
Keep the first test small. Name one main action or change, then describe the details that should stay the same.
3. Review the quote, then inspect the result
The credit amount depends on the model and configuration. Check it before submission; return to your creation to compare, reuse or edit a completed result.
Examples and prompt guides
A scene with generated sound
Compare a simple visual scene with its requested environmental sound.
What to check: Listen for rain and check whether it matches the visible scene; inspect unwanted speech, music or jumps.
Prompt and settings
A cobalt blue ceramic cup on a wooden windowsill while gentle rain falls outside. A locked camera, one continuous close-up. Quiet indoor ambience and soft rain tapping on the window. No music, no voices, no words or logos.
5s · 1080p · 16:9
Prompt and settingsInput limits
Text-to-video starts with a written scene and needs no source image or video. Available settings depend on the selected model; review the exact quote before generating.
Common questions
Do I need to pay before trying?
New accounts receive 100 credits. The selected model and settings determine whether that balance covers a run; the quote tells you before submission.
What happens if I lack credits?
The credit dialog shows the required amount, balance and gap. Add credits or review plans, then return to your creation and confirm a fresh quote. Buying credits does not automatically start a generation.
How do I continue after a result?
Keep the original result in your creation. Reuse its inputs to generate another take, or select it as a source for video editing. Compare each version before using it.