IMAGE WORKSPACE
AI image generator & photo editor
Recent creations
01 / TEXT TO IMAGE
Turn a written idea into an AI-generated image
Describe the subject, setting, lighting, and visual style. Use this AI image generator for product concepts, social content, creative portraits, or illustrations, choosing the model and aspect ratio in one workspace. Without a reference photo, the workspace uses text-to-image generation.
Try a specific starting point: a red ceramic teapot on a cream tabletop, soft window light, minimal product photography, no text. Composition and details communicate your intent more clearly than a topic alone. Results vary with the model and prompt and may need another carefully planned iteration.
After submission, follow task progress in the preview panel, then inspect and download the result. When the provider supplies no live percentage, the workspace labels its progress estimate rather than presenting it as measured completion.
Try it in the workspace
02 / IMAGE TO IMAGE
Edit a reference photo with clear instructions
Sign in to upload JPG, PNG, or WebP photos that you own or have permission to edit, up to 10 MB per file. Multiple references are supported within the selected model's limit, and the workspace numbers them in upload order.
Specify what should change and what should stay. For example: preserve the person and clothing in image 1, replace the background with a warm studio, and use image 2 as the lighting and color reference. Check reference order before submitting and avoid vague instructions such as improve everything.
Adding references selects the model's image-editing workflow. The displayed settings and credit cost still apply. Inspect subject details and edits after completion, and download important results to your device promptly.
Try it in the workspace
03 / MODELS & OUTPUT
Choose model settings for the image you need
The workspace offers Nano Banana 2.1, Nano Banana 2 Lite, Nano Banana 2, Nano Banana Pro, and GPT Image 2.5 Sunburst and Flare. Capabilities differ by model; current availability and supported settings are shown in the workspace.
Choose an aspect ratio for your destination: 1:1 for square layouts, 16:9 for landscape content, or 9:16 for portrait content. Resolution tiers and independent quality options depend on the model. Not every model offers the same 1K, 2K, or 4K choices.
Review the credit cost before submission, including any effect of reference count. All image generation requires sign-in and consumes credits, including any signup bonus credits. Signed-in users can review account history in My creations and should download important images for their own retention.
Try it in the workspace
AI image generator: frequently asked questions
How do I use this AI image generator?
Enter a prompt, choose a model, aspect ratio, resolution, and available quality setting, then submit. The preview panel shows task progress and results. Review the credit cost before generating.
Can I edit an existing photo?
Yes. Sign in, upload reference images, and describe what to change and preserve. JPG, PNG, and WebP are supported up to 10 MB each, with reference limits depending on the model.
Does every model support 4K?
No. Resolution and quality capabilities differ by model. Use the options displayed for your selected model.
Is the progress bar an actual completion percentage?
Provider progress is used when available; otherwise the page shows a clearly labeled estimate. A generated image is shown only after task completion is confirmed.
Do I need to sign in to generate images?
Yes. All image generation and reference uploads require sign-in. Generation consumes credits based on your selected model and settings. Signup bonus credits can be used for generation; purchase credits when your balance is insufficient.
How do I save my generated image?
Click Download image after completion to save the file to your device. Signed-in users can also review account history, but provider links and account retention do not replace your own file backup.
Plan your next image
Review credit options, explore the API, or open your saved creations.