Quickstart
Deploy a fully operational AI agent on dedicated cloud infrastructure in under two minutes. No CLI, no Docker, no YAML.
What you'll get
Create your account
Head to lobstack.ai/onboarding and sign up with Google or email. Authentication is handled securely via Google or email magic link — no passwords to remember.
Choose your AI model
Pick Use Lobstack AI for the simplest start — no API keys, usage included in your plan — or bring your own provider key. You can change this anytime from Settings without losing your data.
Popular choices:
- Lobstack 1.2 — Our flagship managed model; routes each message to the best engine automatically (recommended)
- Claude Sonnet 4.5 — Best balance of speed and capability
- GPT-5.2 — State-of-the-art reasoning and vision (Pro+)
Select your plan & region
Pick a plan based on your usage needs. Every plan includes a dedicated cloud server — no shared infrastructure.
- Starter ($29/mo) — 1,000 messages, 2 vCPU, 2 GB RAM
- Pro ($59/mo) — 5,000 messages, all models, all regions
- Performance ($99/mo) — 15,000 messages, API access, webhooks
- Enterprise ($299/mo) — 50,000 messages, 5 agents, SLA
Choose the region closest to your users: US East, US West, EU Central, EU North, or Asia (Singapore).
Deploy
Click Pay & Deploy. Lobstack will provision your dedicated VM, install the agent bridge, and configure your AI model. This takes about 90 seconds.
You'll be redirected to the dashboard where you can watch the provisioning progress in real time.
Start chatting
Once your agent is running, you'll see the chat interface. Your agent can execute terminal commands, read and write files, browse the web, and use any skills you enable.
Try asking it something like:
What's next?
- Enable skills — Connect Discord, GitHub, Gmail, and more
- Configure your agent — Customize its personality with SOUL.md
- Explore models — Compare all 24 supported AI models from 10 providers
- Explore the Sandbox — Watch your agent work in real-time