Self-hosted AI agents: what to run and where to host them
The open-source AI agents worth self-hosting in 2026, what hardware each needs, and the cheapest VPS or GPU cloud that runs them well.
Updated
This guide is being written. Below is the outline and our current price data; test results will be added once our testing is complete.
Quick comparison
| Provider | From | Starting plan | Free tier | Our rating | Link |
|---|---|---|---|---|---|
| HostingerChecked | $2.99/mo | Premium (web hosting)20 GB SSD | No (30-day money-back guarantee) | Not rated | VisitOfficial site |
| Hetzner CloudChecked | $6.49/mo | CX23 (Cost-Optimized, Intel/AMD)2 vCPU · 4 GB RAM · 40 GB NVMe | No | Not rated | VisitOfficial site |
| DigitalOceanChecked | $4/mo | Basic Droplet 512 MiB1 vCPU · 0.5 GB RAM · 10 GiB SSD | Occasional promo credit for new users | Not rated | VisitOfficial site |
| VultrChecked | $2.50/mo | Cloud Compute Regular Performance, IPv6-only 0.5 GB1 vCPU · 0.5 GB RAM · 10 GB SSD | No free tier | Not rated | VisitOfficial site |
| ElestioChecked | $11/mo | Linode 'Micro' (cheapest shown)1 vCPU · 1 GB RAM · 25 GB SSD | 3-day trial with $20 credit | Not rated | VisitOfficial site |
| RunPodChecked | $0.34/hr | RTX 4090 24GB (Community Cloud) | No free credits; small referral bonus after a $10 top-up | Not rated | VisitOfficial site |
What a self-hosted AI agent is
A short, plain explanation: an agent framework running on your own server, calling either an API model or a model you host yourself. Why people self-host (privacy, cost control, always-on automations).
Which agents to run
A table of the popular open-source agents (OpenClaw, n8n AI agents, and others readers search for) with what each is good at and its minimum RAM and CPU.
Do you need a GPU?
- API-backed agents run on a small CPU VPS (2 vCPU, 4–8 GB RAM).
- Agents with a local model need a GPU, or a lot of RAM for small quantised models.
- Rule of thumb table: model size → VRAM needed.
Where to host it: our picks
One pick per situation, backed by our test servers:
- Cheapest VPS that runs an API-backed agent reliably
- Best one-click setup (Hostinger’s agent and n8n templates, Elestio)
- Best for a local model (hourly GPU from a GPU cloud)
Setup in 15 minutes
Step-by-step on the top pick, with security basics: firewall, updates, secrets, backups.
Costs per month
Worked examples: VPS + API tokens vs a rented GPU. Link to the AI token cost calculator.