PC Praha AI workstation with a single graphics card

AI workstation PC Praha – Ryzen 9, 1 GPU

RTX 5080 16 GB / 128 GB DDR5 / 2 TB NVMe
174 290 Kč
Skip to product information
PC Praha AI workstation with a single graphics card
  • PC Praha AI workstation with a single graphics card
  • AI workstation – photo of the build
  • AI workstation – component close-up
AI workstation PC Praha – Ryzen 9, 1 GPU

PC Praha · AI and GPU servers

AI workstation PC Praha – Ryzen 9, 1 GPU

Specifications by configuration

Graphics cards
1 GPUs
Graphics memory
16–96 GB
Processor
AMD Ryzen 9 9950X
Memory
128–192 GB
Form factor
Workstation tower
144 041 Kč excl. VAT
174 290 Kč incl. 21% VAT
Graphics cards
RAM
Storage

Assembly and testing: usually 5–7 working days. We often have the server in stock and ship it straight away – we always confirm the date individually.

Graphics cards
RAM
Storage
  • 24-month warranty, claims handled directly by us
  • We help with configuration and software
  • We can run it in our data centre on request

What the build includes

Graphics cards
1× RTX 5080 16 GB 16 GB VRAM in total
Processor
Ryzen 9 9950X · 16C
Motherboard
ASUS ProArt X870E-Creator
Memory
2× 64 GB DDR5 128 GB in total
Storage
2 TB NVMe
Power supply
1650 W power supply
Chassis
Workstation tower
Cooling
AIO liquid CPU cooler
Assembly
Assembly, OS installation and 72-hour stress test

The price above is for the complete server including assembly. We are happy to quote a different CPU, network card or chassis.

A quiet workstation with one powerful graphics card for the office or studio. Image and video generation, AI app development and smaller language models right on your premises – no cloud and no monthly fees. Choose the card, memory and drive below.

Full description, video and specifications

Which models this build can run

The selected configuration has 16 GB of graphics memory (VRAM). It comfortably fits, for example, these open language models:

  • Mistral Small 3 24B14 GB
  • Qwen3-14B12 GB

The larger model Gemma 3 27B needs roughly 18 GB of VRAM.

Indicative estimate for quantised models (mostly Q4) with headroom for context. Actual memory needs depend on context length and the number of concurrent users.

What we take care of

  • Custom pre-installation

    We pre-install Linux, NVIDIA drivers, CUDA, Docker and the software for your workload (e.g. vLLM or Ollama) – you can start working as soon as you switch it on.

  • 72 hours under load

    Before handover every server runs at full load for 72 hours while we check stability, temperatures and ECC memory errors.

  • Service within 24/48 hours

    We keep key spare parts in stock in Prague. Service visits and part replacement within 24/48 hours.

  • Hosting in our data centre

    No server room? We place the server in our data centre, take care of power, cooling and connectivity, and you access it remotely.

How we work on AI servers

Description

AI workstation PC Praha is the smallest and quietest build in our range of AI computers. A single graphics card sits directly on the motherboard, the 16-core AMD Ryzen 9 9950X is cooled by a liquid cooler and the whole system runs in a tower case next to your desk. It suits designers, studios and developers who want to generate images and video or develop AI applications on their own machine rather than in a rented cloud.

Configuration

Processor AMD Ryzen 9 9950X (16 cores, AM5)
Motherboard ASUS ProArt X870E-Creator
Graphics card 1× of your choice: RTX 5080 16 GB, RTX 4090 24 GB, RTX 5090 32 GB, Intel Arc Pro B70 32 GB or RTX Pro 6000 96 GB
Memory 128 or 192 GB DDR5
Storage NVMe 2 or 4 TB
Power and cooling 1.65 kW power supply, liquid CPU cooler (AIO)
Chassis tower workstation, quiet operation

Which card for what

Graphics memory Example configuration What runs comfortably
16 GB RTX 5080 language models up to ~24B (Qwen3-14B, Mistral Small 3 24B), SD 3.5 Large images
24 GB RTX 4090 models ~27–32B (Gemma 3 27B, Qwen3-30B-A3B), FLUX.1 [dev] at full precision
32 GB RTX 5090, Arc Pro B70 models up to ~36B (Seed-OSS-36B), Mochi-1 or Wan 2.1 14B video
96 GB RTX Pro 6000 Llama 3.3 70B, GLM-4.5-Air 106B, Qwen3.5-122B-A10B (MoE); large image and video models (HunyuanImage-3.0, HunyuanVideo bf16)

Model sizes are indicative (Q4 quantisation unless stated otherwise), based on our configurator. Actual memory requirements depend on context length and the number of concurrent users. For language models the memory of the cards adds up; image and video generation needs the whole model on a single card.

RTX 5090 and RTX Pro 6000 (Blackwell) cards support FP4 and are significantly faster at image and video generation than older generations. Intel Arc Pro B70 has no CUDA – it suits llama.cpp and OpenVINO; we will help you choose the software.

What you get

  • A server built from new components – we assemble and test it in Prague.
  • Pre-installed software for whatever you want to run on the server (e.g. Ubuntu, drivers and CUDA, Docker, vLLM, llama.cpp, Ollama, ComfyUI). It is ready to work as soon as you switch it on.
  • 72 hours of stress testing under full load before handover.
  • 24-month warranty. We keep spare parts in Prague and provide service intervention within 24/48 hours.
  • The server does not have to stand at your premises – you can keep it in our data centre (hosting) and we take care of power, cooling and connectivity.

Operation

At full load the workstation draws roughly 0.6 to 0.9 kW depending on the card, so a standard power socket is enough. If you need more cards or 24/7 operation in a rack, take a look at our Rome and Genoa servers.

Frequently asked questions

Why only one graphics card?
The workstation has the card directly on the motherboard for quiet operation and simplicity. For more cards we offer the Rome, Edge, Genoa and Turin servers with 2 to 16 cards.
Which card should I choose for generating images and video?
The whole model must fit in the memory of a single card. For images 24–32 GB is enough (RTX 4090 or RTX 5090); for large video models we recommend the RTX Pro 6000 with 96 GB.
What software is pre-installed?
As agreed – for example Ubuntu or Windows, drivers, CUDA, Docker, ComfyUI, Ollama or llama.cpp. We will agree in advance on what you will be using.
How long does delivery take?
We build the workstation to order and test it under load for 72 hours before handover. We will confirm the delivery date after you order.

Customer reviews

There are no reviews yet.

Only customers who bought the product from us can leave a review – we send them a review link by e-mail after the purchase.

More AI servers

View all