---
title: "Deploy kwant"
description: "Your own team of AI bots with handoffs, approvals and routines"
category: "AI/ML"
url: https://railway.com/deploy/kwant
---

# Deploy kwant

Your own team of AI bots with handoffs, approvals and routines

**[Deploy kwant on Railway](https://railway.com/template/kwant)**

Machine-readable deploy manifest (JSON, validated by TemplateCI): https://railway.com/deploy/kwant/manifest.json

- **Creator:** My Projects
- **Category:** AI/ML

## Template content

### browser https://www.google.com/chrome/static/images/chrome-logo.svg

- **Source:** alexsergivan/kwant

### kwant https://github.com/alexsergivan.png

- **Source:** alexsergivan/kwant
- **Health check:** /api/health
- **Public domain:** Yes

## Documentation

# Deploy and Host kwant on Railway

kwant is your own team of AI bots in one self-hosted binary: spaces and chats, bots that hand work to each other, approvals before risky steps, routines on a schedule and a verified-task receipt when a check passes. It runs on API keys (OpenAI, Anthropic, OpenRouter, Ollama and more) or on the Claude Code, Codex and Grok CLIs.

## About Hosting kwant

This template runs the released kwant server image (`ghcr.io/alexsergivan/kwnt`, through the project's own `docker/render.Dockerfile`) with a Railway volume on `/data` for its SQLite database, files and memory, plus a private Chromium service that gives the bots a shared browser. The kwant service is the only one with a public domain; the browser is reachable only over Railway's private network. A sign-in token is generated at deploy time: copy `KWNT_API_TOKEN` from the kwant service's Variables tab and paste it on the sign-in screen. Then add an account (an API key) in Settings and start a team.

## Common Use Cases

- An always-on team of AI bots you reach from any browser or your phone
- Scheduled routines and research jobs that keep running when your laptop is closed
- Bots that browse the web for you in a private, server-side browser

## Dependencies for kwant Hosting

- An API key for at least one model provider, added in the app after deploy

### Deployment Dependencies

- kwant repository: https://github.com/alexsergivan/kwant
- Self-hosting notes: https://github.com/alexsergivan/kwant/blob/main/docs/self-host-stores.md

### Implementation Details

The kwant service starts as root only long enough to hand the `/data` volume to uid 10001, then drops privileges, exactly as the upstream Render image does. `KWNT_ALLOWED_ORIGINS` follows the Railway domain, and `KWNT_BROWSER_CDP_URL` points at the browser service on the private network. The Laya router is not included (kwant falls back to keyword routing), which keeps the template light.

## Why Deploy kwant on Railway?

Railway is a singular platform to deploy your infrastructure stack. Railway will host your infrastructure so you don't have to deal with configuration, while allowing you to vertically and horizontally scale it.

By deploying kwant on Railway, you are one step closer to supporting a complete full-stack application with minimal burden. Host your servers, databases, AI agents, and more on Railway.

## Similar templates

- [Chat Chat](https://railway.com/deploy/-WWW5r) — Chat Chat, your own unified chat and search to AI platform.
- [Weaviate](https://railway.com/deploy/weaviate-1) — Vector database with hybrid search and an API key-protected API
- [Möbius Personal AI OS](https://railway.com/deploy/mobius-v01-personal-ai-os) — Self-hosted AI agent that builds and runs apps for you, with memory.

Open this page in a browser: https://railway.com/deploy/kwant
