AI Photo Editing Tools

Each tool is a full working page with a before-and-after example, the exact instruction it uses and tips for good results.

Every tool combines three things: a preset written instruction, an image model chosen for the job and a rule block that tells the model to keep faces, text, framing and every unmarked detail. The model follows these instructions as well as it can, but it is a generative model and can still change details. When you paint an area with the brush, the model sees a mask and Pixmint composites its answer back onto your original through a feathered edge, so pixels outside the brush come from your original (the edge is blended).

When no brush is used (for example in the background remover), the model returns a re-rendered version of the whole image. That handles things like hair and soft edges well, but small details can shift, so always compare with your original before you publish.

Signed-in users get the preset's stronger model: GPT Image 2.5 for most tools and Nano Banana 2 for document shadows. Guests run the tools on the free Seedream 5.0 Flash model, except document shadow removal, which uses Nano Banana 2 for everyone.

• Use a one-click tool when your goal is one of the common fixes: remove a shadow, remove an object, remove a background.
• Use the AI image editor when you want something else: change the color of a jacket, swap a background for a specific scene, add an object.
• Use the AI image generator when you want a new picture rather than a fix to an existing one, with or without reference images.

• Do you have the right to use the photo? Only upload photos you took or may use. See the content policy.
• Is it the original file? Screenshots and heavily compressed copies carry less detail, and the tools cannot add back what is missing.
• Is the subject sharp and reasonably lit? Blur and noise carry through to the result.
• Do you know which tool fits? A shadow on a face is a job for the portrait scene, a stranger in the background is a job for the object remover, and a new scene behind a product is a job for the editor.
• Will you compare the result with your original? Always do. Generative tools can shift small details such as fine text, jewelry chains or skin texture.

Photos are scaled to a 2048 px long side in your browser before upload, and results from free credits are up to 1024 px with a small watermark. Uploaded and generated images are stored for 24 hours for guests and 7 days for free accounts (30 days is planned for paid accounts), and can be deleted earlier from History.

Each tool run costs 15 credits ($0.15) whichever model the preset uses. Failed runs are refunded. The AI image editor costs the credits of the model you pick (5 to 36 credits per edit).

Guests get 3 tool runs and 5 free-model images per day. Signed in, you get 50 credits on first sign-in and 20 daily credits. Free results are up to 1024 px with a small watermark and are for personal non-commercial use. Planned paid plans (not available yet): 500 credits for $4.90, 1,100 credits for $9.90, 2,400 credits for $19.90, or a monthly plan of 1,200 credits for $9.90. Paid plans are planned but not available yet, so nothing can be bought today (see pricing).