First demo of Orca hybrid quantum AI supercomputer
Cette publication existe aussi en Français
UK startup Orca Computing has shown the first successful demonstration of a hybrid quantum/classical algorithm using its photonic quantum computer with Nvidia’s CUDA Quantum technology.
The integration of the room temperature Orca system with CUDA Quantum uses a standard data centre rack to add existing GPU clusters for AI quantum research.
- ORCA Computing to ship photonic quantum computing AI
- UK startup Orca Computing buys US photonics group
“ORCA is committed to driving transformative change and unlocking new horizons in computational science,” said Richard Murray, PhD, Co-founder, and CEO of ORCA Computing. “Our latest demonstration underscores the potential of hybrid quantum-classical computing to address the world’s most complex challenges.”
At the core of the integration is the ORCA PT-1 photonic quantum processor. In this demonstration, the focus is on image generation using a hybrid generative adversarial network (GAN) approach. The output of the quantum processor is fed to neural networks running on GPUs to produce higher-quality data. This demonstration represents just one facet of the capabilities CUDA Quantum helps ORCA Computing achieve.
CUDA Quantum bridges between the quantum processor and GPU components involved in the data generation process with low-latency communication and synchronization.
“Quantum computing has the potential to solve some of the world’s biggest challenges,” said Timothy Costa, Director of High-Performance Computing and Quantum at NVIDIA. “This latest demonstration, using NVIDIA CUDA Quantum in ORCA’s photonic quantum processor, is set to help customers advance their quantum research.”
If you enjoyed this article, you will like the following ones: don't miss them by subscribing to :
