QCGPU (CUDA version)

Forked from QCGPU, which supports OpenCL.

Open Source, High Performance & Hardware Accelerated, Quantum Computer Simulator. Read the research paper.

This branch settings set CUDA as default (state.py has use_cuda=True).

Features:

Written with OpenCL and CUDA. Accelerated your simulations with GPUs and other accelerators, while still running cross device and cross platform.
Simulation of arbitrary quantum circuits
Includes example algorithm implementations
Support for arbitrary gate creation/application, with many built in.

Installing

Install pycuda by:

$ pip install pycuda

Then, install this package:

$ git clone https://github.com/TRSasasusu/qcgpu
$ git checkout feature/cuda-bydefault
$ python ./setup.py bdist_wheel
$ pip install dist/*.whl

You can use this CUDA version QCGPU with Qiskit by Qiskit QCGPU Provider (CUDA version).

For more information read the full installing docs.