Petuum solutions come pre-integrated with NVIDIA technology to accelerate model training, inference and run-time performance out of the box.

Petuum & NVIDIA

How We Integrate

The Petuum development team uses NVIDIA GPU cards and CUDA-X AI libraries, including cuDNN, in solution development. This shortens time for model training, inference, and of course when deployed into customer production. Our library integration takes full advantage of NVIDIA GPU capabilities, delivering maximum throughput for the most complex AI/ML use cases.

Customer Benefits

Petuum integrates seamlessly with NVIDIA systems to deliver top performance on Day 1. No need for any manual tuning, Petuum and NVIDIA customers benefit immediately, allowing for focus on business problems instead of technology implementation. No matter the model or algorithm across NLP, computer vision or other machine learning, Petuum’s NVIDIA integration delivers critical benefits such as higher throughput, unlimited scaling up/out, and high ROI.


NVIDIA is advancing GPU computing for deep learning and AI at the speed of light. It starts with the most advanced GPUs and the systems and software we build on top of them. We create computers and software to bring AI to the edge, from self-driving cars to autonomous robots to medical devices.