ModelBrain: one memory for every AI assistant you use.
Claude, ChatGPT and Cursor share the same memory, kept on your own computer. Tell one assistant something once, and the others already know it.
For macOS, Windows and Linux. Every page here says plainly which parts aren't finished.
Illustrative example, from your own note, Sep 13.
- Shared by every assistantSwitch apps and your memory comes with you.
- Free on your computerNo account, no card, no cap on how much you remember.
- PrivateSync and backup are encrypted on your computer first, so what we store for you we cannot read.
What stays on your computer, what leaves it, and the limits: see Under the hood.
How it works
Install it
Download the installer for macOS, Windows or Linux. There is no sign-up screen. It creates a private folder for your memories on your computer and runs quietly in the background.
Connect your assistants
Claude Desktop is set up for you, and Claude Code and Cursor take one button. ChatGPT connects with a free account.
Forget it's there
Carry on talking to your assistant: "remember this", "what did we decide on pricing?", "forget what I said about the venue". Day to day, you just talk to your assistant.
Free on your computer. Pay only for sync.
Local memory is free, permanently: not a trial, not a limited plan, not a tier with an asterisk. Device sync and team plans are optional paid services, because they involve a computer ModelBrain runs, not one of yours; the storage they use is billed at a flat $0.05 per GB per month.
No account, no card, no licence key, no expiry, and no cap on how much you remember. Commercial use included.
Three things worth knowing
It stays with you
Your memories are files in a folder you own. There's no account, so nothing is sitting on our servers to leak or be sold.
What leaves your computer →We tested it in public
We built a test designed to catch our own mistakes, and published the results, including the nasty cases.
How it was tested →Deleted means deleted
When you delete something, it's deleted. Nothing comes back unless you asked for it to.
Exactly how forgetting works →Under the hood
You don't need any of this to use ModelBrain. It is here for when you want to check our work.
How we tested it, and the results
Any memory tool can list features. The question that matters is whether it brings back the right note, whether it points you at the note it actually used so you can check it, and whether it keeps doing that once you've got a thousand notes instead of ten.
So we built a test designed to catch our own mistakes, including the nasty cases, like two similar projects, a decision you changed your mind about, and a note you deleted. Across that test, it gave the right answer 86% of the time, and when it genuinely doesn't know something, it says so instead of guessing, every single time. It's still labeled beta, not a finished 1.0, and the full numbers are on the recall test results page. We also ran a separate retrieval test on public benchmarks, in a single locked run: see the retrieval benchmarks.
It shows you which note
It holds up as notes pile up
Deleted stays deleted
What leaves your computer
Your memories are files in a folder you own. There's no account, so there's no password, no profile, and nothing sitting on our servers to leak or be sold. Free isn't a trial; it's the whole thing, for one person, on their own machines.
If you later want your memories shared between your laptop and your desktop, or with your team, that's a paid add-on you switch on deliberately. It is available now, and it is never the default.
Recalling a memory No*†
Usage tracking Local log only
Backup or sharing Only if you set it up
*ModelBrain itself never sends it anywhere. Once your assistant has the memory, that's subject to your assistant's own policies. †True for Claude Desktop, Claude Code and Cursor, which talk to ModelBrain directly; ChatGPT connects through a relay that handles it in transit, not stored.
Privacy in detail →The limits, in plain terms
- Your memory folder stays on your computer by default. Memories you share with an assistant are subject to that assistant's own data policies.
- Synced and backed-up memories are end-to-end encrypted: they are encrypted on your computer before they are sent, and we do not hold the key.
- ChatGPT connections pass through our relay in transit and are not stored.
- The memory folder itself is not encrypted by ModelBrain today. Turn on your computer's own disk encryption (FileVault, BitLocker or LUKS). A key stored next to the vault would add nothing to that, so we do not plan one: why.
- You can read the memory file yourself, and delete anything in it whenever you want.
Exactly how forgetting works
"Forget that" takes a mode, and the modes are written down. By default, forgetting a memory you had corrected removes the current version and brings the previous one back as current. You can instead remove just that version, or the whole correction history, and nothing comes back. A memory that was never corrected is simply gone. Ask it to forget a document you added, and the whole document goes, not a random paragraph of it.
Nothing comes back unless you asked for it in a forget. When you change your mind about something, the old version is kept as history you can look at on purpose, rather than competing with the new one behind your back.
A document → all of it
Comes back on its own → never
Optional: answers from your own notes, worked out on your computer
Point ModelBrain at a folder of notes on your computer, and every AI assistant you use can search it, then reason over what it finds.
Most memory tools find notes that look like your question: a ranked pile of snippets that resemble what you asked. Then your assistant has to sift through them and fill in the gaps itself.
ModelBrain works differently: a small AI model runs on your own computer, reads the notes the search turned up, and answers from what they actually say: one answer, with the specific notes it used cited, so you can check it yourself. It's built to tell you when the answer isn't in your notes, instead of guessing.
That's cognitive inference search: a local model that reads your notes and answers from them, with sources: not a keyword match, and not a guess.
Built for questions your notes can answer: facts, decisions, who's who, what was said. It isn't built for judgment calls yet.
Two AI models run entirely on your computer: one that finds candidate notes, one that reads them and answers from them. ModelBrain downloads and manages both: open-weight models, sized for your hardware, run locally via llama.cpp. No API key, no cloud inference, no per-question cost. The search and the reasoning happen on your machine; your assistant sees the question it asked and the answer it gets back, same as with any tool it uses.
ModelBrain picks the reasoning model automatically: a lighter model on modest hardware, a stronger one if you have the memory to spare. Either way, it runs without any setup. Models download once, the first time you turn Reasoning Mode on in the tray; after that, it works offline.
The reasoning model is read-only: it can read your notes to answer a question, but it can't write to or delete your memory. Before an answer goes back to your assistant, ModelBrain checks that the notes it cites exist and share content with what it says.
Using several assistants, or a whole team
A memory that lives inside one app is a memory you lose when you switch apps, and one nobody else can benefit from. ModelBrain sits outside the assistant, so the same memories are there whichever one you open, and the same vault can be shared with the people you work with.
Tell Claude something on Monday, ask Cursor about it on Thursday. Switch tools next year and the memory comes with you, because it was never inside the tool.
"What did we agree with this client?" gets the same answer for everyone, and a new colleague's assistant knows the history on day one. Private memories are excluded from sharing before it's even considered.
Building on it? Docs · Architecture · Retrieval benchmarks · Facts · Comparisons
Claude memory vs ChatGPT memory
Which one lets you take your memory with you: what you can export, in what format, and whether another assistant can use it.
Can ChatGPT use a memory that lives on your computer?
Yes, with the desktop app. On the web it needs a hop that is reachable from the internet, and whether ChatGPT can write depends on your plan.
Chat Export Reader
A free tool that opens a ChatGPT, Claude or Grok export in your browser and searches every conversation. Nothing is uploaded.
AI chat history: read, search, move, delete
One guide to getting your history out of an assistant, finding things in it, using it somewhere else and deleting it.
The numbers behind the recall claims are on the benchmarks page.
Who it's for
People who use an AI assistant most days for real work and are tired of re-explaining the same context every morning. People who keep notes, decisions and client details they'd rather not hand to a company's servers. And developers who want the protocol details: those are on a page of their own, not spread across this one.
What it doesn't do yet
It's a beta on macOS, Windows and Linux. The Linux build is command line only.
It reads text documents: Word, PowerPoint, Excel, PDF, Markdown, plain text. Not scanned pages, which have no text in them to read.
It won't notice on its own that two of your notes contradict each other. You tell it which one is current.
Sharing memories with a team, and sync between your own devices, are both on paid plans.