compute
one stack, from our cluster to the instrument
where does solpera's ai run?
Our instruments read their own experiments, so the AI has to live where the measurement happens. We run one open-weight stack at every scale: a four-node cluster in Hobart for development and heavy analysis, and NVIDIA Jetson AGX computers on board Spectrophage and Atlas. Models, data and experiments stay under our control at every step.
Our four-node NVIDIA DGX Spark system supports open-weight models up to 400B parameters and provides a common inference layer for spectral analysis, 3D reconstruction, agentic workflows and field-system development. Operating the stack locally allows each workflow to be designed around the instrument and governed as part of the programme.
Because the same stack runs from the cluster to the instrument, we can build it for others too: laboratories, institutions and companies that need AI on their own hardware, from one device at the edge to a multi-node system.
| infrastructure | four-node NVIDIA DGX Spark system · Hobart |
|---|---|
| research workloads | spectral analysis · 3D reconstruction · multimodal inference · agentic workflows · simulation |
| operating model | local data and inference · open-weight models |
| hardware stack | NVIDIA DGX, RTX and Jetson AGX · optimised to each use case |
| deployment range | edge hardware · workstations · multi-node systems |