Video Generation

Create marketing videos using AI — from text descriptions or reference images.

Text-to-Video

Describe the video you want:

• "Generate a video of a girl dancing on a beach at sunset"
• "Create a drone shot over a futuristic city"
• "A product showcase of a perfume bottle rotating slowly"

Image-to-Video

Upload a reference image and animate it:

1. Upload your image in the chat
2. Describe the motion you want: "Make the person wave", "Add camera zoom", "Animate the leaves"
3. The agent will use your image as the first frame

Image Modes

Mode	Description
First/Last Frame	First image is the starting frame, second (optional) is the ending frame
Multi Reference	All images serve as style/content references (up to 5)

Video Settings

• Duration — Select 5, 6, 8, 10, or 15 seconds depending on the model, or "Auto" for the model's maximum
• Aspect Ratio — 16:9 (landscape), 9:16 (portrait/mobile), 1:1 (square)
• Resolution — 480P, 720P, 1080P, or 4K depending on the model

Direct Generation Mode

For quick generation:

1. Switch to Video mode using the type selector
2. Optionally upload a reference image
3. Type your prompt and configure settings
4. Click send

Computing Power

Video generation costs more computing power than images. Duration and resolution affect the cost. The estimated cost is shown before generation.