Save your first 250 items free. After that it's $20/month, and everything works the same — there's no locked-off feature, just a bigger library.
250 is roughly where the connections start getting interesting. You'll know by then whether this is for you.
Because ChatGPT answers from everything ever written, and Everett answers from what you've read. You get sources you recognize, positions you can back up, and an answer that's yours rather than the internet's average. Also: Everett tells you when it's reasoning on its own rather than citing you.
Those file your notes. Everett connects your ideas. In those tools a link just means two things touch; Everett works out what kind of connection it is — what backs up what, what sharpens it, what's the same argument from another field — and then puts it in front of you when it matters.
They're good at getting things in and getting highlights back out. Everett does that too, but the point is what happens after: it works out how everything you've saved relates, and uses it to answer questions and test the positions you take.
Not really. A second brain is still storage — somewhere to put things. Everett is built for the part after storage: using what you've got.
It does, for reading what you save, working out connections, and drafting answers. It never decides what you think. Anything it came up with on its own is labeled in the answer, separately from the things you saved and the people you follow.
We spend an enormous of compute connecting your library and we're proudly not venture-backed so we're not burning capital to unsustainably subsidize it. In return, if your library is meaningful, we ask less than $1/day from you to help you think better.
Anything you save to your library: one PDF, one webpage, one note. If you follow anyone (like a substack or a blog) those items don't count unless you explicitly save them to your library.
Nothing you've saved goes away and nothing stops working. You just can't add more until you subscribe.
PDFs, webpages, Substack posts, notes you write yourself, and photos of pages — Everett reads text out of images.
Upload on the web. Drag and drop works, including up to 10 PDFs at once or a CSV with up to 100 web links.
Yes.
Drag and drop a folder of PDFs and they all go in. If there's an app you use and you need a connection to it, reach out.
No. No tags, no folders, no maintenance. Everett files things by what they're about. You can make collections (like a folder but can be shared with others) if you want them, but nothing breaks if you never do.
Saving one thing is useful — you get a summary and it's searchable. The connections need company: somewhere around 30 or 40 items you'll start seeing things you didn't expect, and it keeps getting better from there.
Let's say there's a blog or substack you read. You can "follow" it via Everett and we'll pull those posts into answers when you ask questions (you control what is actually saved in your library; following doesn't automatically add it to your library).
You can also share collections (they're like a folder) and any you follow will be used to answer any questions you ask.
Your library is stored in the cloud, on Digital Ocean. It's yours — nobody else reads it.
Everett only uses open-source models, run with zero data retention. None of your content is used to train any model.
Yes, all of it, any time — one click.
You keep everything. Every item in your library exports in a single click, so nothing is stranded here.