PC Praha Edge AI server – graphics cards on risers, top view

AI server PC Praha Edge – mobile tower, up to 6 GPUs

2× RTX 5080 16 GB (32 GB VRAM) / 256 GB DDR4 ECC / 2 TB NVMe
358 390 Kč
Skip to product information
PC Praha Edge AI server – graphics cards on risers, top view
  • PC Praha Edge AI server – graphics cards on risers, top view
  • Fitting the graphics cards – front view
  • Fitting the graphics cards – rear view
  • PCIe riser cabling to the graphics cards
  • AI server packed for shipping
AI server PC Praha Edge – mobile tower, up to 6 GPUs

PC Praha · AI and GPU servers

AI server PC Praha Edge – mobile tower, up to 6 GPUs

Specifications by configuration

Graphics cards
2–6 GPUs
Graphics memory
32–384 GB
Processor
AMD EPYC 7643
Memory
256–512 GB
Form factor
Rugged tower on wheels
296 190 Kč excl. VAT
358 390 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
2× RTX 5080 16 GB 32 GB VRAM in total
Processor
EPYC 7643 · Milan 48C
Motherboard
ASRock ROMED8-2T
Memory
4× 64 GB DDR4 ECC RDIMM 256 GB in total
Storage
2 TB NVMe
Power supply
2× 2000 W power supply, redundant
Chassis
Rugged tower on wheels
GPU risers
2× PCIe riser
Cooling
CPU cooler and server fans
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.

An AI server in a tower case on castors for an office, workshop, lab or branch without a rack. AMD EPYC platform, two to six graphics cards and redundant power supplies for continuous operation.

Full description, video and specifications

Which models this build can run

The selected configuration has 32 GB of graphics memory (VRAM). It comfortably fits, for example, these open language models:

  • Seed-OSS-36B28 GB
  • DeepSeek-R2 32B24 GB
  • Qwen3-30B-A3B MoE20 GB
  • Gemma 3 27B18 GB
  • Mistral Small 3 24B14 GB
  • Qwen3-14B12 GB

The larger model Llama 3.3 70B / Qwen3-72B needs roughly 48 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 server PC Praha Edge carries the same AMD EPYC platform as our rack-mounted Rome, but in a rugged tower on wheels. It is designed for places without a server room: AI cameras or quality control in production, transcription of meetings and calls at a branch office, or an internal assistant in a company that does not want to send data outside the building. The power supply is redundant for 24/7 operation and the server can easily be moved wherever it is needed.

Configuration

Processor AMD EPYC 7643 (48 cores, SP3 socket)
Motherboard ASRock Rack ROMED8-2T – 2× 10 GbE, IPMI
Graphics cards 2–4× RTX 5080 / RTX 4090 / RTX 5090 / RTX Pro 6000, 2–6× Intel Arc Pro B70 or NVIDIA L4; connected via PCIe risers
Memory 256 or 512 GB DDR4 ECC RDIMM
Storage NVMe 2 or 4 TB
Power supply redundant, for 24/7 operation
Chassis tower case on wheels

How much graphics memory you need

Graphics memory Example configuration What runs comfortably
32–64 GB 2× RTX 5080, 2× RTX 4090, 2× L4, 2× RTX 5090 Seed-OSS-36B, Llama 3.3 70B / Qwen3-72B from 48 GB
96–144 GB 4× RTX 4090, 4× RTX 5090, 4× Arc Pro B70, 6× L4 Qwen3.5-122B-A10B (MoE), Qwen3-235B-A22B (Q3–Q4) with 144 GB
192 GB 6× Arc Pro B70, 2× RTX Pro 6000 GLM-4.6 355B (Q4)
384 GB 4× RTX Pro 6000 DeepSeek-V3.2 / R1 671B (Q2), Kimi-K2 1T (1.58-bit)

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.

For speech transcription (Whisper), voice synthesis or embeddings, energy-efficient NVIDIA L4 cards with 24 GB and 72 W per card are enough. We do not put passive L40 cards into the tower – they belong in a rack with controlled airflow (the Rome or Genoa server).

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

Redundant power means more noise under load – the server is suited to a technical room, a warehouse or an office where noise is not an issue. Depending on the cards it draws roughly 1.1 to 2.8 kW. We can add faster networking (25 to 400 GbE) on request.

Frequently asked questions

How does the Edge differ from the Rome server?
The platform is the same (AMD EPYC 7643, ROMED8-2T), but the Edge is a tower on castors with redundant power supplies for 24/7 operation – no rack needed.
Is the server quiet?
No. Because of the powerful cards and dual power supplies it is loud under load. It belongs in a technical room or a space where noise is not an issue. Our AI workstation is the quiet option.
Which cards are the most energy-efficient?
The NVIDIA L4 – 24 GB of memory at 72 W. Six L4 cards provide 144 GB of graphics memory with low power consumption, suitable for speech transcription, embeddings and smaller language models.
How is the server managed remotely?
The motherboard has IPMI remote management, so you can power on, restart or reinstall the server without physical access.

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