Create: generate video, images and audio

Create is where you generate individual assets with the best AI models.

Categories

• Video — text-to-video and image-to-video.
• Image — photoreal images, art and product shots.
• Voice — lifelike speech in many voices and languages (including your own cloned voice).
• Music and Sound effects — original tracks and foley from a sentence.
• Avatar — talking presenters from a photo.
• Enhance — upscale, restore and remove backgrounds.

Making a generation

1. Pick a category, then a model. Each model card shows what it's best at.
2. Write your prompt. Add a reference image or video if the model supports it.
3. Adjust settings like aspect ratio, duration or resolution.
4. Check the price on the Generate button and start. You can keep working while it runs.

Tips

• Use Elements to keep the same character or product consistent across generations.
• Your plan sets how many jobs can run at once and the maximum resolution.
• Every result is saved to your Library; open it in the editor or share it from there.
• Failed generations are refunded automatically — see Refunds.