Compose mode is used to generate and edit images and videos using reference images, prompts, and AI models.
Compose mode settings
Setting | Description |
Reference images | Images used as visual references for image or video generation. |
Text prompt | Instructions that describe the desired result. |
Model catalog | Selects the AI model that processes the request. |
Aspect ratio | Defines the proportions of the generated image or video. |
Number of output images | Determines how many results are generated (1, 2, 4, or 8). |
Image resolution | Sets the output resolution (1K, 2K, or 4K). |
Image quality | Sets the quality level (Low, Medium, or High). |
Assist | Adds AI assistance that interprets intent and automatically builds prompts. |
Generate an image
Optional: Add reference images.
Select an image on the canvas and click Reference from its menu.
Or upload images from your device.
Enter instructions in the Text prompt field.
Select an AI model from the model catalog.
Select the aspect ratio.
Select the number of output images:
1
2
4
8
Select the image resolution:
1K
2K
4K
Select the image quality:
Low
Medium
High
Turn Assist on or off.
Click Generate.
Generate a video
Select a Start frame and End frame.
Select images on the canvas and click Start frame or End frame.
Or upload images from your device.
Upload media references.
Optional: Enter additional instructions in the Text prompt field.
Configure the video settings:
Select an AI model from the model catalog.
Select the aspect ratio.
Select the number of output images:
1
2
4
8
Select the image resolution:
1K
2K
4K
Select the image quality:
Low
Medium
High
Turn Assist on or off.
Click Generate.
