Creative Suite Annotation is a child extension for Creative Suite that allows users to mark specific areas on images, attach targeted instructions, and send context-aware edit requests directly to the AI image editing engine. It enables more precise, region-specific edits by letting users visually guide the AI through drawn selections and comments.
Requirements
- Prerequisite: Creative Suite must be installed and active on your workspace
- Installtion: Install Creative Suite — Annotation from the Marketplace.
- Enable: Once installed, Annotation Mode availability per plan is managed at Finance → Pricing Plans. The default AI model is configured at Settings → Creative Suite
Activation
Select an image element on your Creative Suite canvas. A contextual toolbar will appear at the top of the selected image with three tabs: Image, Select Image, and Annotate. Click Annotate to enter Annotation Mode. The canvas will switch to annotation view and the annotation toolbar will appear at the bottom of the screen.
Annotation Tools
Once in Annotation Mode, the bottom toolbar provides the following tools:
- Color Picker — Choose the annotation color before drawing.
- Comment Tool — Add a comment tied to a specific point or region on the image.
- Rectangle Tool — Draw a rectangular selection over a specific area.
- Oval/Circle Tool — Draw an oval or circular selection.
- Speech Bubble Tool — Add a labeled comment visually anchored to a region.
- Brush/Pen Tool — Freehand drawing for precise mask painting.
- Lasso Tool — Draw free-form region selections around irregular areas.
Users can create multiple annotations visually, but only one annotation can be selected for submission at a time. The UI clearly indicates the active/selected annotation. Once ready, click Done to confirm and submit the targeted edit instruction to the AI endpoint.
How It Works
After placing and selecting an annotation, users add a prompt describing the desired edit for that region. The system sends the annotated mask along with the prompt to the AI edit endpoint. The AI processes only the selected region, leaving the rest of the image intact. Each annotation submission counts as one AI edit request and deducts credits accordingly. Requests are blocked if the user has insufficient credits.
Admin Controls
Default AI Model: Select the default AI model for annotation submissions at Settings → Creative Suite → Creative Suite Annotations → Default AI Model (e.g., GPT-IMAGE-2).
Vision Model (Replace Text): A separate vision model can be configured for text-replacement operations within annotations (e.g., Auto/recommended).
Plan-Based Access: Enable or disable Annotation Mode per pricing plan via Finance → Pricing Plans.
⚠️ Note: Only one annotation can be submitted per request. If image edit models do not support masks or positional coordinates, annotations are rendered client-side before the final image is sent to the AI edit endpoint.