Quantum Art Achieves 10X Circuit Depth Compression and 30% Error Reduction Using NVIDIA CUDA-Q Platform

Quantum Art has achieved a 10X compression in circuit depth and a 30% reduction in error rates by compiling circuits with its all-to-all connected multi-qubit gates on NVIDIA accelerated computing using the NVIDIA CUDA-Q platform. The company's fully programmable, all-to-all connected multi-qubit gates and advanced compiler serve as a critical resource for implementing circuits at smaller depth, enabling faster runtime and higher performance, thereby shortening the path to commercial applications at scale. Quantum Art's general-purpose compiler automatically optimizes input circuits and substitutes standard operations with efficient multi-qubit gates, consistently delivering order-of-magnitude compression and substantial performance gains.

The improvements, building on the CUDA-Q integration announced earlier this year, were verified in simulation on NVIDIA CUDA-Q quantum-classical integration framework, underscoring the promise of combining Quantum Art's hardware-aware compilation with the NVIDIA accelerated computing ecosystem. Dr. Tal David, CEO of Quantum Art, emphasized that programmable all-to-all multi-qubit gates represent a critical advancement supporting the company's long-term goal of fault tolerant, commercially viable quantum computing. The verification of these results demonstrates the practical implementation of these theoretical advantages in real-world computing environments.

Dr. Amit Ben-Kish, CTO and co-founder of Quantum Art, explained that their compilation technique demonstrates how multi-qubit gates and optimized compilers can compress quantum circuits by an order of magnitude while simultaneously improving performance by 30%. The general-purpose compiler optimizes very large quantum circuits with few multi-qubit gates, and this compilation is verified by using the NVIDIA CUDA-Q platform to operate NVIDIA AI infrastructure. This approach represents a significant departure from traditional quantum computing methods that rely heavily on two-qubit gates and face limitations in circuit depth and error accumulation.

Sam Stanwyck, Group Product Manager for quantum computing at NVIDIA, noted that by allowing researchers to draw on accelerated computing for their work, NVIDIA CUDA-Q is enabling next-generation breakthroughs in quantum computing. Quantum Art's use of CUDA-Q to achieve circuit depth compression and error reduction serves as a clear example of how meaningful performance improvements are being realized by drawing on the latest advances in AI supercomputing. This breakthrough further validates and aligns with Quantum Art's broader roadmap, which centers on scaling multi-qubit gates and reconfigurable multi-core architectures to deliver increasingly powerful quantum systems. The collaboration demonstrates the importance of hardware-software co-design in advancing quantum computing capabilities beyond current limitations.