Help improve this record. Suggest a correction
NVIDIA · indexed by the LabUpstream recipeOpen-weight models on an RTX PRO workstation
NVIDIA’s local gpt-oss support, connected to the details needed for an OEM workstation evaluation.
THE SETUP
An upstream starting point
NVIDIA documents local support for both model sizes on professional RTX PRO workstations. This is a support record; exact runtime arguments and a load-tested configuration still need to be contributed.
Read NVIDIA’s local-model support guidanceWhat supports this record
NVIDIA support statement indexed 10 September 2026. The GPU edition, power limit and OEM configuration require separate validation.
THE KNOWLEDGE AROUND IT