Own US-based AI infrastructure

Buy an NVIDIA-powered GPU server you own. We host it in our Eugene, Oregon data center, run it 24/7, and route AI workloads to it: put it to work from day one, and what you make is paid out in cash.

How rent-to-own works

  1. Buy the machine. Configure an NVIDIA-powered GPU server you own, with 30% down at 8% interest.
  2. We host it. Racked in our Oregon facility: 24/7 power, cooling, and on-site engineering.
  3. Put it to work from day one. We route AI traffic to it: 30% of what it earns is paid to you in cash, and our 70% share retires the balance. Once paid off, the split flips and you keep 70%.

You stay in control: run your own workloads, toggle the machine on or off, and set the hours the network can use it. Monetizing is an option, not an obligation.

Earnings depend on marketplace demand: projections, not guarantees.

Machines

Own it, don't rent it

Cloud rent is a metered bill that never ends and buys you nothing. A machine you own stays yours, and earns whenever you make it available. Hosting it with us costs a fraction of that rent, and on an earning machine it comes out of what the machine makes rather than out of your pocket.

There's a tax difference too: cloud rent is 100% expense, forever. A machine you own is a depreciable business asset. How that applies to you depends on your situation; consult your tax advisor.

Your models run on hardware you own in a US facility you can name: weights, keys, and data never leave machines you control. Hand-built GPU clusters, US-assembled by Andromeda, with N+1 redundant power. Hosted GPUs for Stanford Blockchain, Penn Blockchain, NYU, and Dartmouth Blockchain.

Browse machines · Financing · Docs · Book a 30-minute call

All machines

Workstation RTX 5090

The 5090 workstation build: up to 2× 32 GB GDDR7 on WRX90, RDIMM to 512 GB.

The RTX 5090 is the entry box: 32 GB of GDDR7 Blackwell for 8B–32B models, diffusion, and render work. Ship it as a single-card prosumer tower or a dual-card WRX90 workstation. Buy it in full. At ~$0.50/GPU-hr a single 5090 earns too little for hosting it with us to make sense, so this class is usually shipped to you.

Price

ItemLanded cost
1× NVIDIA RTX 5090 32GB1 GPU$4,500
Threadripper PRO 9955WX (16c)processor · fixed$1,600
WRX90 workstation boardmotherboard / barebones · fixed$1,100
64 GB DDR5 RDIMM1× 64 GB$2,200
4 TB NVMestorage$1,200
Chassiscase + cooling$900
Networking & miscNIC, cabling, sundries$300
Hand assembly: 5 h × $22assembled in Oregon$110
B3IQ Build & AI-Guarantee: 5%$596
Total$12,506

our landed cost, component prices updated Aug 2026 · Hardware at cost. We make 5% on the build, and 15% of what the machine earns once it is running. Managed Hosting is charged on top and netted from your 85%.

Form factor

Form factorPlatformGPUsLead timePrice
Workstation towerRTX 5090 Workstation (WRX90 tower)1–2× NVIDIA RTX 5090 32GB1–2 weeksStarting at $12,506

Own it today

Own it today. Pay in full, and keep 85% of everything it earns.

How ownership works

What it runs

Specifications

ComponentConfigurationDetail
GPU1–2× RTX 509032 GB Blackwell consumer flagship: render, diffusion, and small-to-mid model AI.
VRAM32 GB GDDR7 each · up to 64 GB1.8 TB/s bandwidth for 8B–32B models and image/video generation.
MotherboardWRX90 workstation boardTower board sized for the card count, with the PSU and cooling to match.
CPUThreadripper PRO 9955WX (16c)Enough cores and PCIe lanes to keep the cards fed during render and inference work.
CoolingTuned workstation tower airflowStandard high-airflow tower cooling, monitored 24/7 when hosted.
Power~1.8 kW (1×) to ~2.4 kW (2×)A single PSU sized for the card count.

Hosted in Eugene, Oregon by default. Nothing to invoice while it earns — our 15% and Managed Hosting both come out of what it makes. Our 15% covers offtake: contract sourcing, demand matching, routing, billing, sla, marketplace operations and payouts.