Hetzner Cloud CX32 Pricing 2026: Cost, Specs, and Who Should Use It

Quick Verdict

Hetzner Cloud CX32 is the sweet spot for developers who outgrew a 4GB VPS. For about €8.30/month before taxes, you get 4 shared AMD EPYC vCPUs, 8GB RAM, 80GB NVMe storage, and 20TB of traffic.

The 8GB RAM is the headline: it is the minimum for running Ollama with 7B models, comfortable for an n8n + database stack, and roomy for multiple Node.js apps on one server. If you only need to host a single Node.js API or bot, the smaller CX22 at ~€4.35/month is still the better starting point.

---

Hetzner CX32 Price in 2026

Item	Hetzner CX32
Approx monthly price	~€8.30/month
Billing style	Monthly cloud server billing
vCPU	4 shared vCPU (AMD EPYC)
RAM	8 GB
Storage	80 GB NVMe
Bandwidth	20 TB monthly transfer
Best fit	Ollama/LLMs, n8n stacks, multiple Node.js apps

Prices can vary by region, currency conversion, and VAT handling. Always confirm the final amount in the Hetzner Cloud console before creating the server.

---

Hetzner CX32 Specs: What 8GB RAM Actually Buys You

The jump from CX22 to CX32 is not about making one app faster — it is about what becomes possible:

Spec	What it unlocks
8 GB RAM	Ollama 7B models, n8n + PostgreSQL, or 4-6 small Node.js apps
4 vCPU	PM2 cluster mode with 4 workers, or app + background jobs without contention
80 GB NVMe	Two or three Ollama models (~5GB each), Docker images, logs, and databases
20 TB transfer	Effectively unlimited for API and automation workloads

If you are still choosing a provider tier, start with the broader best VPS hosting for Node.js developers guide first.

---

CX32 vs CX22: When to Upgrade

The most common question. Rule of thumb: CX22 for one thing, CX32 for a stack.

Workload	CX22 (4GB)	CX32 (8GB)
Single Node.js API + Nginx + PM2	✅ Good fit	Overkill
2-3 small apps or bots	✅ Workable	✅ Comfortable
n8n + PostgreSQL + Redis	Tight	✅ Good fit
Ollama with 7B models	❌ OOM-kills	✅ Minimum viable
Next.js SSR + API + database	Tight	✅ Good fit

Already on CX22 and monitoring shows sustained RAM above 75-80%, swap usage, or repeated PM2 restarts? That is the upgrade signal. The CX22 pricing guide has the full monitoring checklist.

---

CX32 vs CX42: Do You Need 16GB?

Plan	Price	vCPU	RAM	Storage	Choose it for
CX32	~€8.30	4	8 GB	80 GB	7B LLMs, n8n stacks, multi-app hosting
CX42	~€17.30	8	16 GB	160 GB	13B models, heavy concurrent workloads

Stay on CX32 unless you specifically need to run 13B+ models, keep multiple LLMs loaded simultaneously, or serve a production app alongside Ollama with real concurrent traffic. Most self-hosted AI setups on a budget land on CX32 and stay there.

---

The Self-Hosted AI Angle: CX32 as an Ollama Box

CX32’s most compelling 2026 use case is running your own LLM endpoint. With 8GB RAM:

• Llama 3.1 8B or Mistral 7B run at 3-10 seconds per response on CPU — fine for automation, APIs, and batch jobs
• One VPS runs the whole stack: Ollama + n8n + a small Node.js wrapper API coexist comfortably
• Flat cost: ~€8.30/month regardless of how many tokens you generate — no per-request pricing

The full walkthrough is in the Ollama VPS setup guide — install, systemd service, and a secured Nginx API proxy. For model RAM requirements and provider comparisons at this tier, see Best VPS for Ollama.

---

Monthly Cost With VAT

Same rules as every Hetzner plan:

Scenario	What to expect
Non-EU or VAT-exempt business	Close to the base ~€8.30 price
EU individual	Base price plus local VAT
Currency conversion	Your card may add a small spread

For budgeting, treat CX32 as a roughly €8-10/month VPS. For accounting, use the final amount shown in the Hetzner Cloud console.

---

Best Alternatives to Hetzner CX32

Alternative	Better when
Contabo Cloud VPS S (8GB, ~$7/mo)	You want the absolute cheapest 8GB and tolerate weaker shared CPUs
DigitalOcean 8GB Droplet (~$48/mo)	You need managed databases and team tooling around the VPS
Hetzner CX22 (4GB, ~€4.35/mo)	You are hosting one Node.js app, not a stack
Hetzner CX42 (16GB, ~€17.30/mo)	You need 13B models or heavy concurrency

For the performance comparison behind these recommendations, see the Hetzner vs DigitalOcean benchmarks.

---

Verdict: Who Should Use CX32?

Use Hetzner CX32 if:

• You want to self-host Ollama, n8n, or another memory-hungry tool.
• You are consolidating several small apps onto one server.
• Your CX22 monitoring shows sustained memory pressure.
• You want the best €/GB of RAM from a reliable EU/US provider.

Do not use CX32 if:

• A single Node.js app is all you run — CX22 does that for half the price.
• You need live chat support or many global regions.
• Your workload pins all cores 24/7 — look at Hetzner’s CPX or dedicated CCX lines instead.

Setup is identical to the CX22: the step-by-step Hetzner setup guide applies to CX32 as-is — just pick CX32 at the server creation step.

---

FAQ

How much does Hetzner CX32 cost?

Roughly €8.30/month before account-specific taxes or regional differences. EU customers pay VAT on top.

What specs does Hetzner CX32 include?

4 shared AMD EPYC vCPU, 8GB RAM, 80GB NVMe storage, 20TB monthly transfer, and full Linux root access.

Can CX32 run Ollama and local LLMs?

Yes — 8GB RAM is the practical minimum for 7B models like Llama 3.1 8B or Mistral 7B. Expect 3-10 second responses on CPU, suitable for APIs and automation rather than real-time chat.

What is the difference between CX22 and CX32?

CX32 doubles the CX22’s resources (4 vCPU / 8GB / 80GB vs 2 vCPU / 4GB / 40GB) at roughly twice the price. Upgrade when you run a multi-service stack or self-hosted AI, not for a single app.

Is CX32 good for Node.js?

Yes, though a single Node.js API rarely needs it. CX32 shines when Node.js shares the server with a database, n8n, or Ollama.

What are the best CX32 alternatives?

Contabo for cheaper 8GB RAM with weaker CPUs, DigitalOcean for managed ecosystem at ~6x the price, or CX42 if you need 16GB within Hetzner.

---

Ready to deploy? Follow the Hetzner step-by-step setup guide (identical for CX32), then add Ollama or n8n on top.