Its own cloud computer
Every tot runs on its own machine — install packages, run apps, build things and share links. Open your tot's computer at any time to inspect its work.
tots are remarkably capable, always-on agents built to handle everything. Each tot gets its own computer, its own memory and its own routines — so it works for you around the clock, and every part of it stays yours.
tots is the open-source version of Dots: the same always-on tot, built in the open — fork it, self-host it, or just name one and start.
Ask me anything, or hand me a task — I can browse the web, use a computer, and remember what matters to you.
tots runs on the model subscriptions and API keys you already own.
Everything on this screen — the model picker, the drawer, the mascot — is a live preview. Click around.
The characters
Every tot gets a look and a temperament you set — shape, colour, eyes, accessory. They breathe, blink, nod off and celebrate while they wait. Poke one and it pokes back.
Open source
Same idea, built in the open: an always-on tot with its own computer, its own memory and its own routines — except you can read every line, fork it, and run the whole thing yourself.
Own it
tots can do nearly anything using their own cloud computer, their own browser, and the apps you've connected.
Every tot runs on its own machine — install packages, run apps, build things and share links. Open your tot's computer at any time to inspect its work.
tots browse the web on their own, follow a thread of research and come back with the answer instead of a list of tabs for you to read.
tots reach your systems through plugins and MCP servers — point them at the tools you already run, or write a new one in an afternoon. Nothing has to be approved by us.
Your tot takes a project and runs with it, even while it's working on several others. Keep bringing new tasks without juggling separate threads.
Your computer and its contents stay separate unless you choose to connect them. Your tot can also be given permission to use your laptop, right alongside you.
tots doesn't resell model access. Your tot runs on the keys you already have — or a model on your own hardware — so you pay your provider, never a markup on top of it.
Extension of you
With a tot that knows your goals and standards, you can take on more without directing every step. Here's what that looks like.
You're a developer working on an app. Your tot watches customer feedback for recurring requests, scopes smaller improvements and bugfixes, builds and tests them, and brings you complete PRs to review with attached videos showing the changes. You can keep building the next major feature while reviewing the fixes and improvements your tot prepares.
You're leading a product launch. Your tot learns your audience, positioning, and creative standards. When the product scope changes, it works through what that means for the story, revises the launch materials, and prepares drafts of asset and doc changes. You return to a launch plan and creative work shaped by the direction you've set, ready for your review.
You're a scientist. Your tot learns your research question and how you evaluate evidence. As new data arrives, it reruns analyses, investigates unexpected results, and updates the figures for your paper. When findings change, it revises the supporting explanation and flags what needs your review. You return from the lab to work that reflects the latest evidence.
You're a sales lead guiding an enterprise deal through technical review. Your tot checks customer requirements and account history against product docs, finds what still needs testing, builds a proof of concept for a key integration, and suggests a solutions engineer who's handled a similar deal. As requirements and test results change, it updates the proposal and test plan and flags open concerns.
You're a content creator. Your tot learns your voice and what matters to your audience. When a new interview transcript arrives, it identifies moments for clips, prepares show notes, and drafts social posts for your approval. It carries your edits across the materials and learns what resonates — giving you more time to create.
New agent
Each agent is its own workspace, with its own computer, chats, notes and routines. What it learns about you is shared with every agent. Give it a name, a body and a mood.
Pip the Indigo Blob with Headphones
Choose a name for your tot
Everywhere
Working with your tot is as simple as having a conversation. You can message or call tots from the web, the desktop app, or your phone. They can also message you with progress, questions, or decisions that need you.
Always in control
tots ship with their own safety and privacy safeguards built into the runtime — you can read them, not just trust them. Set boundaries, follow progress, and stay involved in decisions that need you.
Each tot works on its own cloud computer, while your computer and its contents stay separate unless you choose to connect it. For signing into supported websites, tots can use saved passwords without exposing them to the model.
When you aren't actively working with it, your tot looks for ways to help in the background — we call this proactive research. It does this with tools restricted to read-only, so it can't send messages, change app content, or control your browser or computer. Security safeguards defend against malicious instructions and monitor for harmful behavior; if a safety concern is detected, it can pause or stop the tot's work.
You choose which apps a tot can access, and manage every permission in your own settings — not somebody else's console.
tots start with built-in rules for when to act independently and when to ask for approval. Custom Rules let you allow specific actions, require approval, or block them. Built-in safety requirements always apply, and you can follow progress — including background work — in Activity View and redirect tots as needed.
tots use auto-review to check actions that could affect your accounts or share information against your instructions, Custom Rules, and safety requirements.
This helps determine what work can proceed, what needs approval, and what you must do yourself. Certain sensitive tasks, such as changing a password, always stay with you.
Your tot's work stays yours. On a self-hosted install, chats, memory and files never leave the deployment you control, and nothing in tots ships your workspace off to be trained on. A tot can still make mistakes, so always review consequential work.
For organizations
Your tot works on your behalf; specialist tots take on dedicated responsibilities within your organization.
Your company sets up each tot with its own identity, credentials, and access to the systems it needs to complete its tasks. As it works, the tot adapts to new information and improves with feedback from your team.
We're building on lessons from running specialist agents inside real procurement, invoice processing, email marketing, customer support and commercial contracting workflows.
Self-host the control plane inside your own network, point every tot at your own model endpoints, and keep credentials in the vault you already run. Nothing has to leave your perimeter for a tot to work.
Start with one specialist tot and one workflow. Give it an identity, the tools it needs, and a reviewer — then widen its access as it earns trust. Because the runtime is open, your security team can audit what a tot can actually do.
Get started
tots is the tot bot you can actually own — your name on it, your model choice, your computer, your memory, your routines. Start with one, grow into a team of them.
There is no meter on your tots. Model calls go to whichever provider your keys belong to, so you pay that provider directly — and on a self-hosted install you pay nobody but your host. Add capacity by adding tots, not by buying seats.
Token
The community token for tots, on Solana. This page is the only place the official address is published — copy it from here, never from a reply or a DM.
Coming soon
The Copy button unlocks the moment the address is published here. Until then there is nothing to buy, nothing to sign and nothing to send.
A tot is an always-on agent that gets to know what matters to you, works on your behalf, and takes important work off your plate so you get more of your time and attention back. It has its own cloud computer, learns from feedback over time, and can work toward your goals 24/7.
Yes. You create your primary tot, give it a name, and make it your own — with its own workspace, computer, chats, notes and routines. In tots you can also shape how it looks and how it behaves, and swap the model underneath it at any time.
tots ships with tots Auto, a router that sends quick asks to a fast model and hard ones to a stronger one, and falls back when a provider is at its limit. Point it at the API keys you already have, or at a model running on your own hardware — the tot works the same either way.
tots is the open-source version of Dots: the same idea — an always-on tot you actually own, with its own computer, memory and routines — built in the open so anyone can read it, fork it, self-host it and pick their own models. It's an independent project and isn't affiliated with, endorsed by, or connected to any company that ships a product called Dots.
When you aren't actively working with it, your tot looks for ways to help in the background — we call this proactive research. It uses the apps you've already connected with tools that are restricted to be read-only, so it can't send messages, change app content, or control your computer.
A tot can still make mistakes, so always review consequential work. Auto-review checks actions that could affect your accounts or share information against your instructions and safety requirements, and certain sensitive tasks — such as changing a password — always stay with you.