Early access

Rent GPUs by the day.

Whole machines we build and host in Eugene, Oregon, never a slice of one. Tell us what you need and we match one.

Pick a node.

NodeGPU memoryMemory bandwidthBetween GPUsForm factor
NVIDIA H200 NVL141 GB HBM3e per GPU4.8 TB/s per GPUNVLink4U rack
NVIDIA RTX PRO 6000 Server Edition96 GB GDDR7 per GPU1.6 TB/s per GPUPCIe4U rack
NVIDIA RTX 509032 GB GDDR7 per GPU1.8 TB/s per GPUPCIeWorkstation tower

Market rate: The median hourly price for this GPU across public listings on other marketplaces, polled live. It is context, not a B3IQ price.

Own one: NVIDIA H200 NVL · Own one: NVIDIA RTX PRO 6000 Server Edition · Own one: NVIDIA RTX 5090

The facility

DetailFigure
square feet28,000
carbon-free grid~90%
redundant powerN+1
monitoring and on-call24/7

Tour the facility · Security and compliance

Size a cluster.

Ask for up to 104 GPUs, in whole nodes of 8. Networking between nodes is arranged on request.

Or just call an API.

Serverless inference, for when you want tokens rather than a machine.

OpenAI-compatible

Point your client at our base URL. The rest of your code stays.

We don’t train on your prompts.

We don’t sell them or write them to storage.

How we handle prompts

What do you need?

A request starts with a short call. Your choices come with it. Book a call · Browse machines

Don’t know the size yet?

Tell the team what you’re running. They’ll help you size it. Book a call