ITEN
Home/PaaSLive inventoryUpdated today · CET
PaaS — Platform as a Service • Ventic manages hardware + stack

PaaS Inventory.
Pick your GPU. We bring it online.

Dedicated machines in EU (Milan, Frankfurt, Paris) and US with GPUs available now. OpenAI/Anthropic-compatible, hardware-tuned vLLM, encrypted mesh, auto-shutdown when idle. Purchase via instant wire transfer (SEPA Instant) to IBAN. Deploy in 1 hour during Italian business hours.

Reserve via wire — info@ventic.itChat with technicians on Discord
Deployment
1 hour
IT hours 09–18 CET
Payment
Instant wire
SEPA Instant to IBAN
Invoice
E-invoice + VAT
Via SEPA / IBAN
How to purchasePaaS · Turnkey
  1. 1
    Pick the VM from inventory
    Click “Reserve via wire” — opens an email to info@ventic.it with pre-filled details.
  2. 2
    We confirm availability + IBAN
    Reply during IT hours. We send IBAN for instant upfront wire and exact amount (running + 50% Ventic + VAT).
  3. 3
    Instant wire transfer
    Execute SEPA Instant. As soon as credited, we start. No card, wire only.
  4. 4
    Deploy in 1 hour
    During Italian business hours Mon–Fri 09:00–18:00 CET (excluding holidays). Outside hours → next business morning.
Need help choosing?
Join Discord and talk directly to technicians — fast reply during IT hours, also for custom clusters and multi-GPU.
Open Discord — ventic technicians
Immediate e-invoice. All machines include hardware-tuned vLLM setup, Ventic Agent (mesh + observability) and support.

Available inventory — 8 configurations

Indicative spot/dedicated prices — confirmed via email at booking time. All machines: vLLM, OpenAI/Anthropic compatible, no public IP, auto-shutdown when idle. EU region preferred for GDPR; US on request.

On request
VENTIC-4090-01Entry
RTX 4090 · 24GB
Available
1× NVIDIA RTX 4090
24GB GDDR6X · 1× GPU
CPU
16 vCPU · AMD EPYC
RAM
64GB DDR5
Storage
1TB NVMe Gen4
Network
1 Gbps · 100TB
🇮🇹 Milan, IT · EU
Running cost
€0.68/h + VAT
You pay — PaaS
€0.34/h /h + VAT
50% of running cost — includes server + vLLM setup + Ventic Agent. Invoiced by us.
Best for
Qwen3 8–32B, Llama 3.1 8B — dev, high-throughput inference
Reserve — instant wire
InfoDiscord
IBAN via email after confirmation · E-invoice
VENTIC-4090x2-01Balanced
2× RTX 4090 · 48GB
Available
★ Popular for startups
2× RTX 4090
48GB total · 2× GPU
CPU
32 vCPU · AMD EPYC
RAM
128GB DDR5
Storage
2TB NVMe Gen4
Network
10 Gbps · 100TB
🇮🇹 Milan, IT · EU
Running cost
€1.28/h + VAT
You pay — PaaS
€0.64/h /h + VAT
50% of running cost — includes server + vLLM setup + Ventic Agent. Invoiced by us.
Best for
Qwen3 32B, Mistral Large — dual-GPU tensor parallel
Reserve — instant wire
InfoDiscord
IBAN via email after confirmation · E-invoice
VENTIC-L40S-01Pro
L40S · 48GB
Available
1× NVIDIA L40S
48GB GDDR6 · 1× GPU
CPU
24 vCPU · Intel Xeon
RAM
128GB DDR5
Storage
1TB NVMe
Network
10 Gbps
🇩🇪 Frankfurt, DE · EU
Running cost
€1.05/h + VAT
You pay — PaaS
€0.53/h /h + VAT
50% of running cost — includes server + vLLM setup + Ventic Agent. Invoiced by us.
Best for
Certified inference, vLLM optimized — 70B Q4, long context
Reserve — instant wire
InfoDiscord
IBAN via email after confirmation · E-invoice
VENTIC-L40Sx2-01Pro
2× L40S · 96GB NVLink
On request
2× L40S NVLink
96GB total · 2× GPU
CPU
48 vCPU · Intel Xeon
RAM
256GB DDR5
Storage
2× 1.92TB NVMe
Network
25 Gbps
🇩🇪 Frankfurt, DE · EU
Running cost
€2.08/h + VAT
You pay — PaaS
€1.04/h /h + VAT
50% of running cost — includes server + vLLM setup + Ventic Agent. Invoiced by us.
Best for
70B full precision, RAG pipelines, multi-model
Reserve — instant wire
InfoDiscord
IBAN via email after confirmation · E-invoice
VENTIC-A100-40Pro
A100 40GB SXM4
Available
1× A100 SXM4
40GB HBM2 · 1× GPU
CPU
24 vCPU · AMD EPYC
RAM
192GB DDR4
Storage
1.6TB NVMe
Network
25 Gbps
🇫🇷 Paris, FR · EU
Running cost
€1.49/h + VAT
You pay — PaaS
€0.75/h /h + VAT
50% of running cost — includes server + vLLM setup + Ventic Agent. Invoiced by us.
Best for
Enterprise legacy, CUDA 12.4, training + inference
Reserve — instant wire
InfoDiscord
IBAN via email after confirmation · E-invoice
VENTIC-A100-80Pro
A100 80GB SXM4
Available
1× A100 SXM4
80GB HBM2e · 1× GPU
CPU
32 vCPU · EPYC
RAM
256GB
Storage
2TB NVMe
Network
50 Gbps
🇺🇸 US-East · US
Running cost
€2.05/h + VAT
You pay — PaaS
€1.03/h /h + VAT
50% of running cost — includes server + vLLM setup + Ventic Agent. Invoiced by us.
Best for
Long context 128k, large batch, prefill-heavy
Reserve — instant wire
InfoDiscord
IBAN via email after confirmation · E-invoice
VENTIC-H100-80Flagship
H100 80GB SXM5
Last one left
★ 1 left
1× H100 SXM5
80GB HBM3 · 1× GPU
CPU
32 vCPU · Intel Xeon
RAM
256GB DDR5
Storage
1.92TB NVMe
Network
50 Gbps · InfiniBand
🇮🇹 Milan, IT · EU
Running cost
€3.15/h + VAT
You pay — PaaS
€1.58/h /h + VAT
50% of running cost — includes server + vLLM setup + Ventic Agent. Invoiced by us.
Best for
Flagship — Qwen3 72B, 1,847 tok/s, hardware-tuned vLLM
Reserve — instant wire
InfoDiscord
IBAN via email after confirmation · E-invoice
VENTIC-H100x2-01Cluster
2× H100 · 160GB NVLink
Pre-order
★ 72h lead
2× H100 NVLink
160GB HBM3 · 2× GPU
CPU
64 vCPU · Xeon
RAM
512GB DDR5
Storage
3.84TB NVMe
Network
100 Gbps
🇮🇹 Milan, IT · EU
Running cost
€6.10/h + VAT
You pay — PaaS
€3.05/h /h + VAT
50% of running cost — includes server + vLLM setup + Ventic Agent. Invoiced by us.
Best for
Cluster — 2× H100 per Qwen3 235B, multi-GPU, production
Reserve — instant wire
InfoDiscord
IBAN via email after confirmation · E-invoice
Can’t find the config? Bare metal clusters, multi-GPU 4×/8×, on-prem or custom regions — ask on Discord or via email.
Join Discord — talk to technicians →
Clear terms

Payment, deployment
and what’s included.

No hidden subscriptions, no surprises. All invoiced in Italy, SEPA.

Payment — upfront instant wire
Only SEPA Instant wire to IBAN (instant transfer). IBAN provided via email after availability confirmation. Wire only — no card, no PayPal. E-invoice + VAT issued immediately. Amount = running cost for requested period + 50% Ventic fee.
Deploy — 1 hour in IT hours
Monday–Friday 09:00–18:00 CET (Rome/Milan), excluding national holidays. Requests outside hours → deploy next business morning. Includes: provisioning, vLLM tuning for your GPU, Ventic Agent, endpoint /v1 ready.
What PaaS includes
Scouting and purchasing the server at the best price (spot or dedicated, EU/US), inference stack setup, vLLM hardware tuning, encrypted mesh without public IP, observability, auto-shutdown when idle and wake on request, ongoing support. BYOH instead: same services on your own machine.
FAQ — PaaS inventory
Why wire transfer only?
For transparency and Italian fiscal simplicity. No card fees, immediate traceability, e-invoice. SEPA Instant = credited in seconds, even outside hours.
Can I reserve now and pay later?
No — reservation confirmed only after credit. “Available” machines are first-come with wire. For “On request”/“Pre-order” we send a quote and timeline before the wire.
How long does the server last?
Pay per hour. You can stop anytime — auto-shutdown when idle to avoid consumption. No monthly lock-in. Ask for an estimate via Discord/email.
Support and hours?
Italian technicians during IT business hours on Discord and email (info@ventic.it). Outside hours: reply next morning. Monitoring included.
Can I bring my own model?
Yes — any open-weight on Hugging Face. We recommend Qwen3, Llama 3.3, Mistral, Gemma. We load and tune it for you.
Email — info@ventic.itDiscord — chat now

Daily inventory update. Last check: today, CET. For real-time availability write on Discord.