Autonomous Computer 4
GitHub 1.2k
From €1,314.62Buy now

Own your intelligence.

Never revoked.

Your models don't get deprecated, rate-limited, or pulled. The weights sit on your disk.

Private by physics.

Prompts never leave the building. Nobody trains on your work. Not by policy, by physics.

No per-token bill.

Own the machine and agents can run around the clock. You pay for electricity, not tokens.

Yours to keep.

Fine-tune on what only you know. The model becomes an asset you own, and it sells with the company.

GPUs

Four cards, tuned and burn-tested. We install RTX 5090 or RTX PRO 6000 Blackwell — installed, driver-loaded, and benchmarked as one system before it ships. Plug it in and run.

128–384GB
Total VRAM
419–503TFLOPS
FP32
7,168GB/s
Memory bandwidth
Assembled Autonomous Computer
Detail
Machine
Chassis

Milled from solid aluminum. A solid billet, milled inside and out until only the machine is left. It moves heat like a heatsink and holds true under a hundred pounds of hardware. 15.5 × 15.5 × 16 inch, anodized silver.

66lb
Weight
CNC
Solid aluminum
Machined detail
Machined detail
Riser

A full ×16 to every card. Each GPU reaches the board over its own server-grade PCIe 5.0 riser — shielded, custom-cut, no dropped lanes. Every power run is cut to length and sleeved by hand.

x16
Lanes per card
64GB/s
Per card
Gen 5
PCIe
Custom-cut cabling
Machined detail
Machined detail
Motherboard

ASUS Pro WS W790E-SAGE SE. The hardest choice in a multi-GPU build, solved with a workstation board: every card gets a full ×16, and the BIOS ships pre-tuned so every card links at full width.

7x
PCIe 5.0 ×16 slots
8
DDR5 channels
W790
Chipset
Machined detail
Machined detail
Machined detail
CPU

Intel Xeon w5-3423. The lanes are the point: 112 lanes of PCIe 5.0 means all four cards, the NVMe, and the risers never fight for bandwidth.

112
PCIe 5.0 lanes
12
Cores
4677
Socket
Machined detail
Power

Two supplies, one clean bus. Four cards at full draw pull close to 3000W — cheap power is where multi-GPU builds die. We spec two 2000W supplies with native 12V-2x6 feeding one distribution board. Plan a 240V circuit before it arrives — normal at this scale, better sorted early.

4000W
Power supply
2x
2000W units
240V
Circuit Required
Machined detail
Machined detail
Machined detail
Cooling

Air moved on purpose. Eighteen fans on a single controller, mapped to the airflow path before a panel was cut — not bolted on after.

Machined detail
Machined detail
Machined detail
Memory & Storage

ECC memory, fast NVMe. Six 32GB DDR5-4800 ECC RDIMMs catch the silent bit-flip before it corrupts a run a whole team depends on. The 2TB PCIe 4.0 NVMe loads a 70B checkpoint in seconds, not minutes.

192GB
ECC RAM
2TB
NVMe
7GB/s
Read
Machined detail
Machined detail
Machined detail

GPU

Cards
4x NVIDIA RTX 5090 or 4x NVIDIA RTX PRO 6000 Blackwell
Total VRAM
128 GB (RTX 5090) / 384 GB (RTX PRO 6000)
FP32 compute
419 TFLOPS (RTX 5090) / 503 TFLOPS (RTX PRO 6000)
Interconnect
PCIe 5.0 x16 per card, Gen5 riser cables

Compute

CPU
Intel Xeon w5-3423 - 12 cores, 112 PCIe 5.0 lanes
Motherboard
ASUS Pro WS W790E-SAGE SE - 7x PCIe 5.0 x16
CPU cooler
Heatsink 4677-2UAF8

Memory & storage

RAM
192 GB (6x 32GB DDR5-4800 ECC) - configurable 64-256 GB
Storage
2TB NVMe PCIe 4.0 - configurable 1-8TB, Gen5 on 8TB

Power & cooling

Power supply
2x 2000W, native 12V-2x6 - 4000W total
Wall power
240V circuit required
Fans
18x 120mm, single controller

Dimensions

Size
15.5"L x 15.5"W x 16"H
Weight
66 lbs
Chassis
CNC-machined solid aluminum, anodized silver

Software

OS
Ubuntu, NVIDIA drivers pre-installed
AI frameworks
PyTorch, CUDA, vLLM, SGLang, llama.cpp

Warranty

Full system
1 year

Other

Shipping
2-4 weeks
Autonomous Computer Studio specs

Open-source hardware

The whole machine
is open.

Every CAD file, bill of materials, and BIOS setting — free on GitHub. Fork it, change it, build your own, even sell it. This is how the personal computer began: in the open. We’re doing it again, for AI hardware.

GitHub 1.2k
autonomous-ai / autonomous-computer

2x-5090/ 4x-5090/ 8x-5090/ 4x-6000/

bom/ step_models/ stl-models/ photos/

README.md setup.md

MIT · 147 forks

Clone it. Build it. Sell it. We just want it built.