furybot

Chat with a model that answers to you

A conversational assistant on the same account as the picture tools. Pick a model by the job you want done rather than by parameter count, and switch mid-conversation without losing the thread.

Models named for the work, not the weights

General, code, maths and science, vision. Each one says what it is good at and — more usefully — what it is weak at, so you can pick without knowing what a quantisation level is.

Reasoning stays out of the answer

Models that think out loud produce paragraphs of deliberation before getting to the point. That is kept in its own collapsible block, so the answer reads as an answer.

It will sometimes be confidently wrong

Worth saying plainly rather than in the terms. Language models invent details — names, dates, citations, figures — and state them exactly as they state correct ones. Nothing is checked. Verify anything that matters, and treat legal, medical and financial answers as a starting point for a conversation with a professional.

Web search is off unless you ask for it

Everything else stays on our hardware. Search cannot: answering what happened this week means asking the public internet. So it is a per-message choice, with a notice at the moment you make it, and the answer cites what it found.

Use it from your own tools

The API is OpenAI-compatible, so VS Code, Neovim, Zed and your own scripts work with a settings change and no plugin from us.

Common questions

Is chat included on the free tier?
Yes. Free and paid accounts both include it, drawing on the same compute allowance as the picture tools — so a quiet week of images means more conversation, and the reverse.
How much can I use?
Your account page shows roughly how many more replies are in this week's budget, estimated from your own recent conversations rather than an average nobody matches.
Can it be wrong?
Yes, and confidently so. It invents details and presents them like facts. Check anything that matters.
Is my conversation used for training?
No. It is not sent to any third-party AI service and is not used to train anything. Transcripts are encrypted where they are stored and are yours to delete.

How your files are handled

Everything is processed on hardware we own and operate — your prompts and photos are never sent to a third-party AI service. Uploads are deleted as soon as the job finishes, and results are erased automatically after a short retention period. Download anything you want to keep.

Be furious. Be ferocious. Express yourself boldly.

Private AI chat that runs on our own hardware · Furybot