What you can do
- Generate an image from a description, in square, landscape or portrait formats. OpenAI models can return a transparent background, which is handy for stickers, icons and cut-out product shots.
- Remove or replace a background. One click fills in a prompt that keeps the subject and drops the rest.
- Remove an object and let the model fill the gap.
- Enhance a photo: exposure, white balance, noise and sharpness, without changing the content.
- Restyle an image as a painting, sketch or any look you describe.
Every result can open in PhotoCraft as a new document, be downloaded, or go back through the panel for another round of edits. To add it to a document you already have open, copy it inside PhotoCraft (Select › All, Edit › Copy) and paste it there as a new layer.
How it connects to PhotoCraft
- In the editor, click AI tools in the top bar. The panel opens on the right; the canvas resizes to fit.
- Open the settings (gear icon), choose a provider and paste your API key. The key field links to the provider’s key page.
- To edit your work, use File › Export › Export As… or Layer › Quick Export as PNG in PhotoCraft. The export still downloads as normal, and a copy appears in the panel as the image to edit.
- Pick a preset or write what should change, then press Apply edit (or ⌘/Ctrl + Enter).
- Choose Open in PhotoCraft. The result opens as a new document; from there, Select › All, Edit › Copy, switch to your PSD and Edit › Paste to add it as a layer.
Supported providers
| Provider | Default model | Get a key |
|---|---|---|
| OpenAI | gpt-image-2.5-flare | platform.openai.com |
| Google Gemini | gemini-3.1-flash-image | aistudio.google.com |
| OpenRouter | google/gemini-3.1-flash-image | openrouter.ai |
| Custom (OpenAI-compatible) | any model your API serves | your own endpoint |
OpenRouter gives one key access to many image models, including Gemini, GPT Image, FLUX, Seedream and Qwen Image. The custom option works with any service that implements OpenAI’s /images/generations and /images/edits endpoints and accepts browser requests.
Privacy: what goes where
| Data | Where it goes |
|---|---|
| Your API key | Stays in your browser. Sent only to the provider, in the request header. |
| Prompt and the image you choose to edit | Sent from your browser straight to the provider you picked. |
| Everything else you do in PhotoCraft | Never leaves your device. |
| photocraft.run servers | Serve the static editor files. They receive none of the above. |
The provider’s own terms and data policy apply to what you send them. Large exports are scaled to 2048 pixels on the long side before upload to keep requests fast and within provider limits.
No key? Use PhotoCraft’s local tools
For a plain cut-out you may not need a model at all. Select a photo layer and use Remove Background in the Properties panel, or Select › Subject and refine with Select and Mask. Both run on your device for free. See how to remove a background in PhotoCraft.
FAQ
Do I have to pay to use the AI tools?
You pay your AI provider for what you use, at their normal API prices. photocraft.run charges nothing and adds no markup, because the request never passes through us.
Where is my API key stored?
In your browser only. By default it lasts for the current tab (sessionStorage). Tick “Remember my key on this device” to keep it in localStorage instead, and use “Forget all keys” in the panel settings to delete it.
Which model should I pick?
For OpenAI, gpt-image-2.5-flare is the fast everyday model and gpt-image-2.5-sunburst is tuned for precise edits. For Google, gemini-3.1-flash-image is quick and inexpensive and gemini-3-pro-image follows complex instructions better. “Load models” in the panel lists everything your key can use.
Can the AI change my layered PSD directly?
No. Image models work on flat pictures, so the panel takes an exported PNG, JPEG or WebP and gives back a new image. Open it in PhotoCraft, copy it into your PSD as a new layer, then mask and blend it like any other layer.
Why does an edit fail with a CORS or network error?
The custom provider option needs an API that accepts requests from browsers. OpenAI, Google Gemini and OpenRouter all do. Some self-hosted or regional gateways don’t; use one of the built-in providers instead.
Does PhotoCraft itself have AI features?
PhotoCraft has local, non-generative helpers such as Select Subject and Remove Background, which run on your device without any key. Generative features are not part of PhotoCraft; the AI panel is something this site adds around the editor.