Deploy Rakazo Suite

Rakazo persistent cloud agent (Grok Bot Alternative). Non forked.

Deploy Rakazo Suite

Just deployed

Just deployed

/var/lib/postgresql/data

Just deployed

Deploy and Host Rakazo on Railway

Rakazo is an open-source, self-hosted alternative to Grok Bot for deploying and managing persistent AI teammates. Unlike closed SaaS platforms, Rakazo ensures you retain full ownership of your data, chat histories, model discretion, and agent memory. It also supports collaborative team spaces where members can share team bots and computers or deploy completely private, isolated bots.


About Hosting Rakazo

Hosting Rakazo provides a decoupled, full-stack architecture: a React 19 web frontend, a Hono and oRPC API service that applies Prisma database migrations on boot, a dedicated Graphile Worker background engine to poll and execute tasks asynchronously, a managed PostgreSQL database, and a remote cloud sandbox provider for virtual machine execution.

This Railway template provisions and interconnects the complete stack out of the box using Railway's private networking.

Why Deploy Rakazo on Railway?

  • Total Data Ownership: All message history, credentials, files, and bot memories remain in your own private PostgreSQL database and infrastructure rather than on a third-party platform.
  • Privacy-First & Ultra-Low-Cost Models: Pair your self-hosted setup with open-source models like GLM 5.3 Flash via zero-data-retention (ZDR) providers. This keeps inference costs negligible while ensuring your entire pipeline—from model execution to persistent storage—stays strictly private.
  • Team Collaboration & Private Bots: Organize workspaces into teams where members can share persistent Team Computers and collaborative bots, or run isolated private bots with granular privacy boundaries.
  • 100% Official Upstream Architecture: Built directly from the official upstream repository and container releases without forks, mono-container compromises, or stripped background workers.
  • Frictionless Sandbox Setup: Pre-configured with Daytona as the default sandbox provider—allowing you to start testing with a free tier that doesn't require a credit card upfront—while remaining fully pluggable with providers like Box / boat.dev and E2B.
  • Resource Optimization: Set the Web and API services to Railway's serverless sleep mode so your stack pauses when idle and wakes instantly upon traffic.

Common Use Cases

  • Self-Hosted Grok Bot Alternative: Run persistent AI teammates with native computer-use capabilities (browser, terminal, desktop) while maintaining full control of your infrastructure.
  • Multiplayer Team Workspaces: Deploy shared bots that an entire engineering or product team can collaborate with across web and messaging channels (Slack, WhatsApp, etc.).
  • Private & Sensitive Tasks: Allow team members to run isolated private bots equipped with their own virtual machines for confidential workflows.
  • Autonomous Coding & Research: Direct agents to clone repositories, run test suites, execute bash commands, and browse web interfaces in stateful environments.
  • Background Workflows & Scheduled Routines: Run periodic monitoring jobs and automated cron routines safely via the dedicated Graphile Worker.

Dependencies for Rakazo Hosting

  • PostgreSQL 16: Managed datastore for chat histories, team permissions, credentials, and the background job queue.
  • Sandbox Provider API Key: Daytona (default), Box / boat.dev, or E2B.
  • Model / LLM Provider API Key: OpenRouter (recommended default to easily run models like GLM 5.3 Flash with zero data retention) or any provider connected directly via the UI (OpenAI, Anthropic, xAI, etc.).
  • Integration Provider API Key: Composio allows you to integrate your agents with 1,500+ apps and services. You can also use one of the other integration providers.

Quick Setup

  1. Deploy the Template: Railway will provision the Postgres, Web, API, and Worker services.
  2. Set Your Keys:
    • DAYTONA_API_KEY: Add your API key from Daytona (or set SANDBOX_PROVIDER=box with your BOX_API_KEY from boat.dev).
    • Model / LLM Provider: Use your API key from OpenRouter to use most any model. Other model providers are also available.
    • Integration Provider: Use your API key from Composio to integrate your agent with 1,500+ apps and services. Other integration providers are available.
  3. Launch: Open your generated Railway web domain, register your admin account, and create your first AI teammate.
  4. Optimize: Under service Settings → Deploy, configure the web and api services to use serverless sleep

Template Content

More templates in this category

View Template
Chat Chat
Chat Chat, your own unified chat and search to AI platform.

okisdev
116
View Template
stella
Self-host stella with web, API, Postgres, Redis, and object storage.

Jan Kubica
7
View Template
Hermes Agent | OpenClaw Alternative with Dashboard
Self-Hosted Hermes AI Agent for Telegram, Discord & Slack

codestorm
79