Affiliate links on this page — rates explained
🌊
Best for beginners

DigitalOcean

Easiest setup of any cloud provider. $200 free trial credit included.

Starting at$12/mo
LocationsNYC, SFO, AMS, SGP, LON, FRA, TOR, BLR, SYD
SLA99.99%
🎁 $200 credit for 60 days
Get started on DigitalOcean → See current pricing ↗

We earn 10% of customer spend for 12 months if you sign up.

Prices shown are community-verified as of April 2026. Click the provider link above to confirm current rates — pricing changes without notice.

PlanRAMCPU / GPUPriceOllama use
Basic 2GB 2GB 1 vCPU $12 Phi-3 Mini (3B only) Get plan →
Basic 4GB Recommended 4GB 2 vCPU $24 7B models (Q4_K_M) Get plan →
Basic 8GB 8GB 4 vCPU $48 14B models (Q4) Get plan →
Premium 16GB 16GB 8 vCPU $96 34B models (Q4) Get plan →

Community benchmarks

Measured by community members running real workloads. Numbers are tokens/second on the listed model.

18 tok/s
Qwen 2.5 7B (Q4_K_M)
Basic 4GB
First token ~1s. Good for chat.
28 tok/s
Llama 3.2 8B (Q4_K_M)
Basic 8GB
Fast first token ~700ms.
16 tok/s
Qwen 2.5 14B (Q4_K_M)
Premium 16GB
Smooth 14B on premium CPUs.

Detailed pros & cons

What's good

✓ $200 free credit — risk-free trial

The most generous trial of any provider we tested. At $24/mo for 4GB, you get 8 months free. Plenty of time to validate before spending real money.

✓ Best documentation in the cloud industry

DigitalOcean's tutorials library has thousands of guides, including Ollama-specific walkthroughs. Non-technical users can follow along step by step.

✓ 9 global regions

New York, San Francisco, Amsterdam, Singapore, London, Frankfurt, Toronto, Bangalore, Sydney. Pick the one closest to your users.

✓ Strong API for automation

Full REST API with Terraform provider. If you're building a product on top of your AI stack, DigitalOcean's API is stable and well-documented.

✓ Managed databases alongside your AI

Need a Postgres database or Redis cache next to your Ollama server? DigitalOcean's managed services are easy to add from the same console.

Watch out for

✗ 50–75% more expensive than Hetzner

4GB Basic at $24 vs Hetzner CX22 at €5.89. The $200 credit masks this — once it expires, it's a significant cost difference.

✗ No GPU instances at standard tier

GPU Droplets exist but are enterprise-only and expensive. For GPU inference, use RunPod instead.

Best for

First-timers who want documentation, support, and a $200 free trial. US-based users who need low-latency US data centres.

Not for

Budget-conscious users after the trial credit expires. Hetzner is 50%+ cheaper for the same specs.

Ollama setup guide for DigitalOcean

Step-by-step from account creation to first model response. Takes 15–20 minutes.

1

Create your account & claim the credit

Sign up via our referral link to get $200 free credit valid for 60 days. No catches — it covers 8 months of the 4GB Droplet plan.
2

Create a Droplet

Click 'Create' → 'Droplets'. Choose:
• Ubuntu 22.04 LTS (best Ollama compatibility)
• Your closest region (New York or San Francisco for US users)
• Basic 4GB plan ($24/mo)
3

Add your SSH key

Upload your public SSH key (~/.ssh/id_rsa.pub) for passwordless login. If you don't have one:
ssh-keygen -t ed25519 -C 'your@email.com'
4

Connect and update

ssh root@YOUR_DROPLET_IP
apt update && apt upgrade -y

Takes ~2 minutes. Always update before installing anything.
5

Install Ollama

curl -fsSL https://ollama.com/install.sh | sh

Ollama installs as a systemd service. Check it's running:
systemctl status ollama
6

Pull a model

For code completion:
ollama pull qwen2.5-coder:7b-instruct-q4_K_M

For chat + RAG:
ollama pull llama3.2:8b-instruct-q4_K_M
ollama pull nomic-embed-text
7

Set up the firewall

DigitalOcean Console → Networking → Firewalls → Create Firewall.
Add Inbound Rule: TCP, Port 11434, Source = Your IP.
Assign the firewall to your Droplet.
Never expose Ollama publicly without authentication.
8

Connect your editor or app

Continue.dev in VS Code: Ollama base URL = http://YOUR_IP:11434
AnythingLLM: Server URL = http://YOUR_IP:11434
Open WebUI: OLLAMA_BASE_URL = http://YOUR_IP:11434

Community setups

8GB Droplet + Llama 3.2 8B + AnythingLLM
28 tok/s
"$200 credit covered my entire trial. RAG pipeline on documents works great for my team."
@startup_founder

Ready to get started?

New accounts get $200 credit for 60 days. Enough to test your full setup risk-free.

Get started on DigitalOcean →