Please follow these instructions.
The easiest way to get started with CUDA Quantum is via the public Docker images. These images are available for x86_64
(or AMD64
) and aarch64
CPU architectures.
ghcr.io/nvidia/cuda-quantum:latest
To pull the image, you will need to install docker and then run docker pull <image_name>
.
For instructions on how to run the CUDA Quantum container, refer to this webpage.
Make sure to add --gpus all
to the docker run command to expose all available GPUs to the container
CUDA Quantum programs run natively via backend-extensible circuit simulators. The most performant of these require an NVIDIA GPU (e.g. V100, A100, H100, A6000, A4000, etc.). If you do not have access to such a GPU (e.g. on your Macbook), then you will not be able to target these backends. If you have access to a remote workstation with an NVIDIA GPU that you can access during the hackathon, that would be best.
A --target <target-name>
flag can be specified at compilation for C++
and at runtime for Python
, which is a combination of the desired platform and simulator / QPU.
To get additional information on the simulators and backends, go to TARGETS.md.