AI video generator

Generate AI video from a prompt or an image. Short ads, product clips, and social video — then keep editing in chat.

How it works

1

Describe the shot

Camera, subject, and motion. Or upload a still to animate.

2

Generate a clip

Short-form clips for ads, Reels, and product loops.

3

Edit the clip

Extend, recaption, or restyle in the same chat.

Before & After

Popular Tools

Community Examples

Models

Credits per generation. No subscription. See pricing.

FAQ

What is an AI video generator?
An AI video generator creates short video from a text prompt or from a still image. On ImageGPT you describe the shot, generate a clip, then keep chatting to extend, recaption, or restyle it.
Can I do text to video and image to video?
Yes. Text to video makes a clip from words. Image to video animates a photo you upload or just generated. Both live in the same chat.
How long are the videos?
Clips are short-form — typically a few seconds per generation, which is what ads, Reels, and product loops need. You can extend or chain shots in chat.
Which video models can I use?
Use models such as Kling, Veo, MiniMax, Seedance, and Wan from one place. Pick a model or leave Auto on.
Can I make ads and product videos?
Yes. Teams generate product clips, UGC-style ads, and social videos, then add captions in a workflow if they want burned-in subtitles.
Do I need an account to start?
You can try the AI video generator in chat without a card. Video uses more credits than stills; extra credits unlock more models and volume.