Your own private GPT, on hardware you own
A private ChatGPT-style service on NVIDIA GB10 Blackwell hardware: 128 GB VRAM runs Llama 3.3 70B at full precision for 10 to 20 concurrent users. Prompts and documents never leave your network.
Pick your unit
Both units run the same GB10 Blackwell chip with 128 GB VRAM. Delivery within Riga is included.
NVIDIA GB10 Blackwell
€6975.00 one-time
In stock: 50
- GB10 Blackwell
- 128 GB VRAM
- 4 TB NVMe
- Cluster support
- Runs Llama 3.3 70B full precision
- 10 to 20 concurrent users
ASUS Ascent GX10 128GB
€5752.50 one-time
Low stock: 6
- GB10 Blackwell
- 128 GB VRAM
- 1 TB NVMe
- Single-node only
- Runs Llama 3.3 70B full precision
- 10 to 20 concurrent users
Hardware prices exclude 21% VAT. Delivery within Riga included; elsewhere by arrangement.
We make it a working service
Hardware alone is not a product. Setup turns the box into a private AI service; the managed plan keeps it fast, patched and current.
Setup
€5000.00 one-time
- Unboxing and provisioning at your site or our DC
- Model deployment on vLLM or llama.cpp
- OpenAI-compatible API and chat UI on LAN/VPN
- Optional RAG: Qdrant/pgvector, SharePoint and more
- Hardening: API keys, audit log, rate limits
- Backup and disaster-recovery plan
Managed operations
from €500.00 /mo
- 24/7 monitoring: GPU load, VRAM, queue, latency
- OS and CUDA patching in a monthly window
- Quarterly model refresh (Llama 4, Qwen 3...)
- Nightly RAG re-index and tuning
- User and access management
- 4 hours of engineer time every month
Why run your own LLM
Actually private
Prompts, documents and embeddings stay on a machine you own. Nothing is sent to a third-party AI provider.
Predictable cost
One hardware purchase instead of per-token bills that grow with every user you onboard.
Your site or ours
Run it in your office on LAN, or colocate it in our Riga data centre with VPN access for your team.
Standard integrations
The OpenAI-compatible API means existing tools, SDKs and plugins work without code changes.
Invoices and contracts do not belong in public AI
Pasting client invoices, contracts or HR documents into ChatGPT, Claude or any other public AI hands personal data to a third party, often outside the EU. Under GDPR that needs a legal basis and a processing agreement, and NIS2 makes you answer for your suppliers. A private GPT removes the problem: nothing leaves hardware you control.
GDPR: data stays yours
Prompts, documents and embeddings are processed on your own machine, in your office or in our Riga data centre. No transfer to a third-party AI provider and nothing to explain to the regulator.
NIS2: suppliers under control
The EU directive requires you to manage vendor risk. One box you own is a shorter supplier list than a foreign AI API, and we operate it as an EU provider with the paperwork to match.
Confidential by design
Invoices, contracts, medical and HR records work in chat and RAG without ever leaving your network, so your team gets AI help without leaking client data.
How it works
- 1
Size the workload
We talk through users, models and data sources, and you pick the unit and where it will live.
- 2
We deploy
Hardware arrives, the model goes live behind an API and chat UI, RAG connects to your documents.
- 3
Your team uses it
Staff chat with company knowledge privately; the managed plan keeps models and indexes fresh.
Our own Tier 3+ data centre in Riga
Your services run on hardware we own and operate in Latvia, under EU jurisdiction. A second live region in the Netherlands, Dubai planned for 2026.
- N+1 redundant power and cooling
- Own BGP network with redundant uplinks
- 24/7 on-site engineers
- GDPR and EU data residency
Private GPT questions
Can we use it with documents that contain personal data?
Yes, that is the point. Invoices, contracts and HR files are processed on hardware you own, so no personal data is transferred to a third-party AI provider. That keeps the processing inside your GDPR perimeter and your NIS2 supplier list short, unlike pasting the same documents into a public AI service.
Which models can it run?
The 128 GB VRAM comfortably runs Llama 3.3 70B at full precision, and smaller local LLMs like Qwen or Mistral with room to spare. Quarterly refreshes under the managed plan bring in new releases as they ship.
Is the API really OpenAI-compatible?
Yes. We expose the standard chat-completions API on your LAN or VPN, so libraries and tools built for OpenAI endpoints work by changing only the base URL and key.
Can it answer from our internal documents?
Yes. The optional RAG setup indexes sources like SharePoint into Qdrant or pgvector, and the nightly managed re-index keeps answers current.
Where does the hardware live?
Either on your premises or colocated in our Riga data centre with VPN access for your team. In both cases you own the unit and every byte of data on it.
Do we have to take the managed plan?
No. Setup alone hands you a working, hardened service. The managed plan, from 500 EUR per month, is for teams that want monitoring, patching and model refreshes done for them.
How much does a private GPT server cost?
The hardware is a one-time purchase; current unit prices are listed on this page and exclude 21% VAT. Setup is 5000 EUR one-time and the optional managed plan starts at 500 EUR per month. There are no per-token fees, so costs stay flat as usage grows.
How many people can use it at the same time?
One GB10 unit with 128 GB VRAM serves 10 to 20 concurrent users on Llama 3.3 70B at full precision. If your team outgrows that, the NVIDIA GB10 unit supports clustering, so you add nodes instead of replacing hardware.
What is the difference between the NVIDIA GB10 and the ASUS GX10?
Both run the same GB10 Blackwell chip with 128 GB VRAM, so LLM performance is identical. The NVIDIA unit ships with 4 TB NVMe and cluster support; the ASUS Ascent GX10 has 1 TB NVMe and runs single-node only. Delivery within Riga is included for both.