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.
Two 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.
Milled from solid aluminum. Not a bent-steel box — a billet, milled inside and out until only the machine is left. It moves heat like a heatsink; the triangular cutouts keep it stiff without dead weight. 12.5 × 12.5 × 16 inch, anodized black.
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.
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.
Intel Xeon w5-3423. The lanes are the point: 112 lanes of PCIe 5.0 means both cards, the NVMe, and the risers never fight for bandwidth.
2000W that doesn't flinch. Under full load the box pulls up to 1600W — and cheap power is where multi-GPU builds die. We spec 2000W with native 12V-2x6, feeding clean power to every card. The result is boring: it just runs.
Air moved on purpose. Six fans on a single controller, mapped to the airflow path before a panel was cut — not bolted on after.
.jpg)
.jpg)
.jpg)
ECC memory, fast NVMe. Error-correcting memory catches the silent bit-flip before it corrupts the job you left running overnight. The NVMe loads a 70B checkpoint in seconds, not minutes. Need more? Configure up to 256GB RAM and 8TB Gen5 NVMe.
.webp)
How big is your team?
vLLM • FP8 weights + KV • 1K prompt • 512 outputQwen3.6-27B, per user (tok/s)
Grey: Mac Studio M3 Ultra single user, published run. Blue: this box on vLLM FP8, per user while serving 4 and 24 users at once.
Source:Community benchmark
Runs these models
Language6- Qwen3.6 27B
- Qwen3.6 35B A3B
- Gemma-4 31B
- Gemma-4 26B A4B
- Laguna-S 2.14-bit GGUF
- Laguna-XS 2.1
Image3- Qwen Image 2512
- Qwen Image Editing 2511
- FLUX.2-dev
Video2- Wan 2.2
- MiniMax H3
Quantized builds noted per model. What fits depends on context length and KV cache.
GPU
Compute
Memory & storage
Power & cooling
Dimensions
Software
Warranty
Other
The lineup
Three builds.
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 1.2k2x-5090/ 4x-5090/ 8x-5090/ 4x-6000/
bom/ step_models/ stl-models/ photos/
README.md setup.md
MIT · 147 forksClone it. Build it. Sell it. We just want it built.