Memory

Close a conversation, open a new one, and the assistant you were working with is gone. Not busy, not distracted — gone. It has no idea who you are, what this project is, or that the two of you spent an hour yesterday deciding not to use that library.

This is not a flaw someone forgot to fix. A language model has no memory between conversations. It reads what is in front of it and answers. When the conversation ends, there is nothing left to remember with.

So Semantix does the remembering instead. The model still forgets; the app doesn't. Before your message reaches the model, Semantix looks for what it already knows about this project and puts the relevant parts in front of the model along with your question. The model isn't recalling anything. It's being handed notes.

That distinction is the whole feature, and everything below follows from it.

Because of that: it only knows what was written down#

Notes have to exist before they can be handed over. Memory doesn't quietly absorb everything you type — you decide what's worth keeping, in two ways.

Ask for it in conversation. Tell the assistant to remember something and it can write it down itself: "remember that we're on Postgres 16, not 15." It can also look things up on request — "what do you know about how this project handles auth?"

Hand it a whole conversation with /sleep. At the end of a session that mattered, /sleep reads the conversation back and keeps what's worth keeping, so the next one starts knowing it.

→ Filing a conversation with /sleep

Because of that: it belongs to a project, and you choose which one#

Notes about your API server are noise in your game engine. Memory is scoped: each project has its own, and a conversation draws on one of them.

Usually that's the project you have open, and you never have to think about it. When you do want to think about it — reaching into another project's memory from here, say — the model picker's Memory row is where you change it.

→ What it recalls as you type

Because of that: it lives somewhere, and that somewhere isn't your laptop#

Your code, your conversations and your settings sit in ~/.semantix/ on your machine. Memory does not. It's kept on the Semantix server against your account, which is how it follows you to a second machine — and it's the reason Local only mode switches it off completely.

That's a real trade and you should make it deliberately rather than discover it later.

→ What's stored, and where

The two switches#

Memory is off until you turn it on, and there are two separate controls. They're easy to confuse, so it's worth being precise about which does what.

Settings → Privacy → Semantix Memory decides whether. This is consent. It's off by default, and Local only overrides it — with Local only on, memory stays off no matter what this checkbox shows.

Settings → General → Memory decides how much. Once consent exists, this is where you shape what happens: a master switch and three smaller ones, all on by default. The three shape what gets recalled automatically. The master is broader than it looks — turning it off closes the automatic recall and takes the remember/recall abilities away from the assistant, so it can no longer be asked either. /sleep is the exception: it keeps working.

What this is not#

Memory is not a transcript. It doesn't store your conversations so you can read them back — that's what conversation history is for, and it's a separate thing kept locally.

It's also not automatic understanding. A memory that was never written is not recalled, and one written badly is recalled badly. The assistant is working from notes, and notes can be wrong.

→ What it recalls as you type · Filing a conversation with /sleep · What's stored, and where