Quantum application development for multi-processor systems using CUDA-Q
summary
CUDA‑Q is an open‑source toolchain that compiles annotated C++ and Python quantum kernels through an MLIR/LLVM pipeline into QIR, then into backend-specific streams (NVVM/PTX for GPUs) and backend adapters for QPUs; kernels may embed classical code which is either executed on CPUs/GPUs or compiled away. The stack already runs fast simulators (state‑vector and tensor‑network), integrates with partners like Quantum Machines, exposes Docker images and NVIDIA Quantum Cloud, and targets real-time compute and CUDAQ-toolchain workflows for persistent classical–quantum memory.
AI-written summary · about
chapters (14)
- 0:00Quantum app development intro
- 1:05Quantum-HPC integration
- 3:57Classical needs for quantum
- 5:43CUDA-Q hardware abstraction
- 7:49Hybrid quantum-classical systems
- 10:17Real-time compute for quantum
- 15:12Future of quantum apps
- 18:03Quantum software progress
- 23:43CUDA-Q architecture
- 36:35Split compilation model
- 38:45CUDA-Q toolchain
- 40:32CUDA-Q benchmarks
- 42:16CUDA-Q getting started
- 42:39Q&A
related talks
21mCUDA-QX and Artificial Intelligence for Quantum Computing
16mqTPU: Hybrid Tensor Networks for Quantum-Classical Acceleration
20mQuantum Computing as a Service- Cloud-based and automized QML Solu
21mExtending Modern C++ for Heterogeneous Quantum-Classical Computing
46mFueling the Quantum Application Era with the Cloud
19mQuantum Virtual Machines
18mIntegrating Quantum Computers in High-Performance Computing Environments
21mHybrid Quantum Computing for AI