Choose the starting input that matches the creative material you have.
Start from a written scene description when you do not need an existing image or video to guide the result.
Current workflowAnimate one still image, with an optional end frame on compatible models, when the source composition should anchor the clip.
Animate an image →Transform source footage or combine supported references when motion, pacing, identity, style, or sound should guide the result.
Transform a video →Model availability changes. The editor is the source of truth for the options and credit estimate available before each generation.
| Capability | What to expect | Basis |
|---|---|---|
| Starting input | A written scene and motion prompt | Workflow definition |
| Model, duration, framing, resolution, and audio | Use the choices currently shown in the editor | Live editor controls |
| Output | A newly generated video; exact prompt matching is not guaranteed | Provider guidance |
Everything you need to know before you start creating.
It turns a written scene description into a short video. Your prompt can specify the subject, action, scene, camera movement, pace, and mood.
Start with subject and action, then add the setting, camera move, lighting, and pace. Simpler motion directions usually produce more stable first passes.
Yes. The controls shown for the selected model determine the available duration, framing, and output quality.
Choose based on the controls and output you need. Each model exposes its currently supported durations, aspect ratios, resolutions, and optional features in the editor. You can reuse the same prompt across models and compare the results.
Some models can generate audio with the video, while others return video only. Sound availability is model-specific, so check the controls and model description shown in the editor before generating.
It works especially well for social clips, product moments, creative previews, visual experiments, and early storyboards.
Text to Video creates the scene from a written description without requiring a starting visual. Image to Video begins with an uploaded still image, using its subject and composition as the visual anchor for the generated motion.
Not always. Results are generative, and complex actions, readable text, exact logos, or several simultaneous movements may vary. Start with one clear action and refine one instruction at a time.
Imagine Scenes shows the estimate before generation. The amount depends on the selected model and its supported settings, such as duration, resolution, and optional audio. Review the estimate again after changing a control.
Select a secure payment provider to continue.
Create amazing AI-generated videos and images
Create a new account with Google or GitHub to claim them. Unused credits expire 30 days after signup.
By continuing, you agree to the terms and privacy policy.