Deploy VibeCody Anywhere

Your own always-on AI coding assistant — zero DevOps required.

VibeCody runs on anything from a $35 Raspberry Pi to a full cloud instance. Pick your platform, run one command, and you’re live in under 5 minutes.

Choose Your Platform

Cloud Platforms — Always-On, Team-Ready

Best for teams, CI/CD integration, and 24/7 availability. Estimated cost: $5–60/month.

Platform	Setup Time	Monthly Cost	Free Tier	Guide
AWS	10 min	$15–60	12-month free tier	Deploy →
Google Cloud	10 min	$10–50	$300 credit	Deploy →
Azure	10 min	$15–55	$200 credit	Deploy →
Oracle Cloud	10 min	$0	Always-free ARM	Deploy →
DigitalOcean	5 min	$12–48	$200 credit	Deploy →
Linode / Akamai	5 min	$12–48	—	Deploy →

Best value: Oracle Cloud’s always-free tier gives you 4 ARM cores + 24 GB RAM — enough for VibeCody + a 7B local model at $0/month.

Desktop — Personal Workstation

Best for individual developers who want VibeCody integrated into their daily workflow.

Platform	Setup Time	Always-On	GPU Acceleration	Guide
macOS (MacBook / Mac Mini)	2 min	✅ launchd	✅ Metal	Install →
Linux (Ubuntu, Fedora, Arch)	2 min	✅ systemd	✅ CUDA/ROCm	Install →
Windows	3 min	✅ Service	✅ CUDA	Install →

Edge / Homelab — IoT & Self-Hosted

Best for privacy-first setups, smart home hubs, and always-on personal assistants.

Platform	Setup Time	Local Models	Remote Access	Guide
Raspberry Pi 5 (8 GB)	5 min	Up to 7B	Tailscale / Cloudflare	Install →
Raspberry Pi 4 (4–8 GB)	5 min	Up to 3B–7B	Tailscale / Cloudflare	Install →
Raspberry Pi 3 (1 GB)	10 min	TinyLlama (cloud recommended)	Tailscale / Cloudflare	Install →

Homelab tip: A Mac Mini with Apple Silicon is the best desktop “server” — 18 GPU cores run 13B models at full speed with zero fan noise.

One-Command Setup

Every platform has a one-liner. Here’s the fastest path for each:

macOS / Linux

curl -fsSL https://raw.githubusercontent.com/TuringWorks/vibecody/main/install.sh | sh
vibecli

Windows (PowerShell)

irm https://raw.githubusercontent.com/TuringWorks/vibecody/main/deploy/windows/setup.ps1 | iex

Docker (Any Platform)

docker run -p 7878:7878 ghcr.io/turingworks/vibecody:latest

Cloud (AWS Example)

git clone https://github.com/TuringWorks/vibecody.git
cd vibecody/deploy/aws
./setup.sh

Raspberry Pi

curl -fsSL https://raw.githubusercontent.com/TuringWorks/vibecody/main/deploy/raspberry-pi/setup.sh | sh

Always-On Mode

VibeCody can run as a persistent background service, just like myclaw.ai — but self-hosted and free.

Platform	Service Manager	Command
macOS	launchd	`vibecli --serve --port 7878 --provider ollama`
Linux	systemd	`vibecli --serve --port 7878 --provider ollama`
Windows	Windows Service	`vibecli --serve --port 7878 --provider ollama`
Docker	docker-compose	`docker compose up -d`
Cloud	Managed (auto)	Always running after deploy

Once running, access VibeCody from anywhere:

# Local access
curl http://localhost:7878/health

# Remote access (via Tailscale)
curl http://vibecody.tailnet:7878/health

# Gateway integrations (Slack, Discord, Telegram, etc.)
vibecli gateway enable slack
vibecli gateway enable discord
vibecli gateway enable telegram

Tiers

Like myclaw.ai’s Lite/Pro/Max plans, VibeCody offers tier presets for resource allocation — but you control the hardware and pay only infrastructure costs.

Tier	vCPU	RAM	Storage	Best For	Cloud Cost
Lite	2	4 GB	40 GB	Chat + small agent tasks	$5–16/mo
Pro	4	8 GB	80 GB	Agent loops + local 7B models	$20–33/mo
Max	8	16 GB	160 GB	Multi-agent + local 13B models	$40–66/mo

Resource allocation is configured through your infrastructure provider’s settings (instance size, container resource limits, etc.) rather than through VibeCody CLI flags.

How It Compares

Feature	VibeCody	myclaw.ai (OpenClaw)
Cost	Free (self-hosted) + infra costs	$16–66/month
AI Providers	23 (local + cloud)	Limited to what OpenClaw supports
Code Intelligence	7 review detectors, 550+ skills, LSP	General-purpose
Deployment Targets	12 platforms	Managed cloud only
Raspberry Pi	✅ Full support	❌
Desktop App	✅ VibeUI (196+ panels)	❌
Open Source	✅ MIT License	Wraps open-source, hosted service
Always-On	✅ All platforms	✅ Cloud only
MCP Integration	✅ Full	✅ Via OpenClaw
Agent Autonomy	Agent loop + multi-agent teams	Single agent
Gateway	18 messaging platforms	~10 integrations
Privacy	Your hardware, your data	Managed servers
Customization	Full config + plugins + rules	Limited
CI/CD Integration	GitHub Actions, K8s, Docker	Limited
GPU Acceleration	Metal, CUDA, ROCm	Depends on plan

What’s Next

Use Cases — See 80+ things you can do with VibeCody
Quickstart — Your first chat in 60 seconds
Configuration — Deep-dive on all settings
Tutorials — Step-by-step walkthroughs

Easy Setup