Uploading Images
This guide explains how to add images for annotation in the Playground: supported formats, the ways you can upload, and what to expect before you run a task.
Summary
- Where: In the Playground, inside the Image & Prompt card.
- Formats: PNG, JPEG (JPG), and WebP. Other file types are not accepted.
- Ways to add: File picker (click), drag-and-drop, or paste from the clipboard.
- Before processing: You need at least one image. You can remove or replace it until you submit the task.
For the full flow (upload → prompt → process → result), see Quick Start.
Supported formats
Only image files of these types are accepted:
| Format | MIME type | Notes |
|---|---|---|
| PNG | image/png | Common for screenshots and graphics. |
| JPEG / JPG | image/jpeg, image/jpg | Common for photos. |
| WebP | image/webp | Modern format; good compression. |
- If you pick, drop, or paste a file that is not one of these, it will not be added. No error is shown in the UI; only supported images are used.
- File extensions like
.png,.jpg,.jpeg, and.webpare typically matched to the types above. The app uses the file's MIME type to decide.
How to add an image
You can add an image in three ways. All use the same Image & Prompt card on the Playground.
1. File picker (click)
- Click the image placeholder (empty box with an icon) or the + (plus) button next to the controls.
- A system file dialog opens. Choose one image (PNG, JPEG, or WebP).
- The image is added and a small preview appears in the card. If you already had an image, it is replaced by the new one.
2. Drag and drop
- Drag an image file from your computer (or another app) and drop it onto the Image & Prompt card.
- Only the first supported image from the drop is used. If you drop multiple files, only one supported image is added; if you already had an image, it is replaced.
- Dropping unsupported file types (e.g. PDF, GIF, BMP) does nothing visible; only PNG, JPEG, and WebP are accepted.
3. Paste from clipboard
- Copy an image to the clipboard (e.g. Copy from a screenshot tool, or copy a region/image in an editor).
- Click inside or near the Playground card so it has focus, then press Ctrl+V (Windows/Linux) or Cmd+V (Mac).
- If the pasted content is a supported image type, it is added (or replaces the current image). Other clipboard content is ignored.
What you see after adding
- A preview of the image appears in the top-left of the card (thumbnail).
- You can remove it by clicking the × on the preview; the card returns to the empty state and you can add another image.
- You can replace the image at any time by adding a new one (click, drop, or paste); the previous image is replaced.
- In the current Playground UI, the card holds one image at a time for that task. Add your image, then optionally set a prompt and options, and click Submit to process (see Quick Start).
Tips
- Screenshots: Use your OS or tool shortcut to capture a region/window, then paste (Ctrl+V / Cmd+V) into the card for a fast workflow.
- Quality: Very large images may take longer to upload and process. If you hit timeouts or slowness, try a smaller or compressed version.
- Unsupported types: If nothing happens when you add a file, check that it's PNG, JPEG, or WebP. Convert or export to one of these if needed.
What happens when you submit
When you click Submit (or use Ctrl+Enter / Cmd+Enter):
- A task is created on the server (after checking your account and credits).
- Your image is uploaded to the server as the task's reference image.
- The AI job is submitted. The view switches to the task/chat screen where you see status and, when ready, the annotated result.
Upload and processing are described in more detail in Quick Start and Understanding Results.