Photo AI

Your AI photo editing studio

Settings

Done

Image generation & generative edits

Used for text-to-image, generative edits ("add a sunset sky"), AI backgrounds and studio headshots. Everything else — adjustments, background and object removal, enhancement, upscaling, OCR and photo search — runs on your device.


Free with no key: each image you create or edit counts as one of your free daily uses in this app. With an AIV key (Settings → AI provider) it's unlimited.

Anonymous use is throttled (about one image every 15 seconds) and may add a small watermark.

Start AUTOMATIC1111 or Forge with --api --cors-allow-origins=* (or your site's address). Runs fully on your computer.

Please don't create deceptive images of real people, and only make AI headshots of yourself or people who agreed.

On-device models

Downloaded once on first use and cached by your browser.


AI assistant (text & vision)

Choose where your AI assistant runs. Keys stay in this browser and are only sent to the service you pick.


Free AI from AI Verse, ready to use with no key. Every app includes 20 free daily uses of AI. For unlimited, full-strength AI, get an AIV key.

Full-strength AI from AI Verse, recommended for full use. One key works in every AI Verse app. Get an AIV key


Free tier, recommended for testing. Create a key at console.groq.com/keys.


Use your own key from any OpenAI-compatible service. A few services block requests from web pages — Test connection will tell you.


Run a model on your own computer with Ollama or LM Studio. Nothing leaves your machine. Works on the computer running the model.


Setup

Ollama

  1. Install from ollama.com and download a model, e.g. ollama pull qwen3:8b (7–8B models or larger work best for planning and email).
  2. Allow this site to talk to Ollama by setting OLLAMA_ORIGINS to , then restart Ollama.
    Mac: launchctl setenv OLLAMA_ORIGINS "" · Windows: add it under System → Environment Variables · Linux: OLLAMA_ORIGINS= ollama serve

LM Studio

  1. Load a model, open the Developer tab and start the server.
  2. Turn on Enable CORS in the server settings.

If your browser asks to let this site access devices on your local network, allow it.

A small model downloads once into this browser and runs on your graphics card — fully private, and it works offline afterwards. Needs Chrome or Edge with WebGPU. Answers are simpler than cloud models. Food photos are still recognized on-device; AI physique reviews need a vision model from another provider.


Your data

Photos, versions and settings are stored only in this browser. An image is sent to a provider only when you use a cloud feature (generation, generative edit, describe).