High-Performance AI/ML Workstation Service

The buyer requires a high-performance AI/ML workstation service for a critical laboratory asset. The service must support offline large language model fine-tuning, optimization, and execution, as well as deep learning neural networks. The system must be compatible with existing laboratory models, pre-compiled deep learning frameworks, and proprietary software packages, specifically requiring NVIDIA Grace/Blackwell hardware and CUDA architecture. An AI Developer Toolkit, including standard NVIDIA AI Development Tools, CUDA Toolkit, NVIDIA drivers, and parallel computing libraries, must be preinstalled and tested for immediate local model training. The contract period is for five years. This bid is set aside for total small businesses, and eligible vendors must be onshore (USA organizations only). Performance of the work will be offsite. The buyer is seeking proposals.