NVIDIA CUDA-QX

Quantum researchers and developers across a wide range of domains, from quantum error correction to hybrid solvers, use GPU programming to accelerate their applications. This demands highly optimized, domain-specific libraries. NVIDIA CUDA-QX, built on top of CUDA-Q™, is a collection of libraries and tools for accelerating research and development toward useful accelerated quantum supercomputing.

Get Started

Quantum Computing Stacked Diagram

CUDA-QX Libraries

CUDA-QX Libraries are built on top of CUDA-Q, NVIDIA’s open-source, hardware-agnostic platform for accelerated quantum supercomputing, and also released open-source on GitHub. The CUDA-QX libraries provide optimized implementations of key quantum primitives, from quantum error correction to hybrid algorithms, enabling developers to easily leverage the CUDA-Q platform.

CUDA-Q QEC

Including GPU-accelerated decoding primitives, Google’s stim stabilizer simulator, and extension points in CUDA-Q for custom decoders, CUDA-Q QEC is the foundational toolkit for any error correction researcher.

Learn More About CUDA-Q QEC

CUDA-Q Solvers

Run prebuilt optimized kernels for VQE, ADAPT-VQE, QAOA, GQE, and more to get the most performance out of today’s hardware.

Learn More About CUDA-Q Solvers

Latest Product News

Introducing NVIDIA CUDA-QX Libraries for Accelerated Quantum Supercomputing

Learn how the CUDA-QX libraries can accelerate development of hybrid applications ranging from quantum error correction to chemical simulation.

Learn More About CUDA-QX Libraries

Google Uses CUDA-Q to Simulate Quantum Device Physics

Accelerated computing enables prohibitively complex simulations to improve quantum hardware performance.

Read Press Release

Get Started With CUDA-QX Today.

Get Started