Home / Lab / Infrastructure

The Lab Behind wAIve

Every response you get from wAIve runs on metal we own. Seven machines and 624 GB of GPU memory: a fully local AI lab in Oshkosh, Wisconsin, with no cloud in the loop.

7machines
624 GBGPU memory
15GPUs
24/7one pool

Inside the lab

Two NVIDIA DGX Spark systems with 128 GB of unified memory each, a Blackwell RTX PRO 5000, an RTX 5090, and eleven GeForce RTX 3090s, orchestrated as one pool by our own fleet-control software, running 24/7.

That's where wAIve lives: large language models, image generation and editing, and a voice agent that speaks in real human voices, all running on our own hardware.

Big-lab compute. Hometown scale.

HardwareGPU memory
2× NVIDIA DGX Spark256 GB (unified)
11× GeForce RTX 3090264 GB
RTX PRO 5000 Blackwell72 GB
GeForce RTX 509032 GB
Total624 GB

This is the stack wAIve runs on.

And it's how we size the hardware we build for clients: we run it here first.

Start with a free exploration call.

Tell us what you're trying to do and what's in the way. If AI fits, we'll show you where. If it doesn't, we'll say that too.