The model on your device answers by default. When you want a bigger one, bring your own key: Claude, OpenAI, OpenRouter, Gemini, Groq, or Ollama and LM Studio running on your computer. Your browser talks straight to the provider. Fogar has no server, so there is no copy of your key or your conversations anywhere but here.
Free. Open source. You pay your provider, if anyone. Fogar never sees the bill or the key.
Add a model in the LLM Chat widget and pick where it lives. Each preset fills in the endpoint and a sensible model; Fetch models asks the endpoint what it actually serves.
| Preset | Talks to | Key |
|---|---|---|
| Anthropic | Claude, through the Messages API.api.anthropic.com · reads images | console.anthropic.com |
| OpenAI | GPT models.api.openai.com/v1 · reads images | platform.openai.com |
| OpenRouter | Hundreds of models behind one key.openrouter.ai/api/v1 | openrouter.ai |
| Gemini | Google’s models, through their OpenAI-compatible endpoint.reads images | aistudio.google.com |
| Groq | Very fast open models.api.groq.com/openai/v1 | console.groq.com |
| Ollama | Whatever you pulled, on this computer.localhost:11434/v1 | None |
| LM Studio | Whatever you loaded, on this computer.localhost:1234/v1 | None |
| Custom | Any server that answers /v1/chat/completions. | If it needs one |
Profiles are shared by every widget that picks a model. Add one once, use it everywhere below.
Open its settings, choose a preset, and paste your key. For Ollama or LM Studio there is no key; the server just has to be running.
keys stored in this browser · sent only to their own endpointFogar asks the endpoint what it serves, so the list is what your account or your machine actually has, not a list we typed. Start typing to pick one.
/v1/models · Ollama lists what is installedChrome asks for access to that one address when you save or fetch, never at install. Pick the model in the widget’s header; the conversation is kept for your next tab.
optional permission · one origin at a timeSettings has its own cloud endpoint for the ask box, recipes, the bookmark finder, and attached files. It speaks the OpenAI format: OpenAI, OpenRouter, Groq, or Ollama at localhost all work.
On any answer, Dig deeper can put the same question and conversation to your cloud model, or back to the one on your device. The better answer replaces the first.
The Writing widget can distill and rewrite with any profile, so Claude can write as you. How the voice works.
Mark a profile as reading images and the Notebook can hand it a photo of a handwritten page. The Canvas can use any profile to draw what you describe.
The rest of Fogar is a new tab that answers from a model running in your browser. See what leaves in each mode.
No. There is no Fogar server. The new tab sends each request straight to the endpoint you configured, with your key, and to nowhere else. The network ledger in Settings counts every request the page makes, by host.
Yes. In the LLM Chat widget, add the Anthropic preset and paste a key from console.anthropic.com. Fogar talks to the Messages API directly from your browser. The Writing and Canvas widgets can pick the same profile. The ask box’s cloud endpoint speaks the OpenAI format, so for Claude there, use OpenRouter.
Yes, with no key. The presets point at localhost:11434 for Ollama and localhost:1234 for LM Studio, and Fetch models lists what is installed or loaded. The Settings cloud endpoint takes Ollama too, so the ask box can use it. The server has to be running.
Fogar cannot know your endpoint in advance, so it lists none at install. When you save or fetch models, it asks Chrome for access to that one origin, and only that one. If you decline, the profile is still saved, and Fogar says the permission was not granted.
In Chrome’s extension storage on this device, unencrypted, like any extension setting. It is sent only to the endpoint it belongs to. Remove the profile, or clear the field in Settings, and it is gone.
Yes. Each LLM Chat widget keeps its conversation in extension storage, so it is there on your next tab. It holds the most recent 60 messages or 30,000 characters, whichever comes first. New chat clears it.
Fogar is on the Chrome Web Store. One click to add it, and it updates itself from there. Then open a new tab.
Free, from the Chrome Web Store. Brave, Arc, and Edge install from it too.
On your phone, or not on Chrome? The same Fogar runs as a web app at app.fogar.ai. Open it once with a connection, add it to your home screen (Safari: Share, then Add to Home Screen; Chrome: the menu, then Add to Home screen), and set it up from there; after that it works offline. The widgets that need Chrome’s own APIs, Sessions, bookmarks, Inbox, Feed, Jira, and Agenda, stay on the desktop.
Loaded the zip before the listing went live? The store copy has a different extension ID, so your data will not follow on its own. Three steps move it over.
Todos, notes, recipes, widgets, sessions, and settings go into one file.
It opens on your next new tab, empty.
Once it looks right, remove the old copy in chrome://extensions.
Can't use the store? The same build is here as a zip: unzip it, turn on Developer mode in chrome://extensions, and Load unpacked. It will not update itself.