CUDA-Q

[westurner]

How can tequila implement support for CUDA-Q?

NVIDIA/cuda-quantum; CUDA-Q:

• src: https://github.com/NVIDIA/cuda-quantum
• docs: https://nvidia.github.io/cuda-quantum/latest/
• docs: https://nvidia.github.io/cuda-quantum/latest/using/examples/introduction.html
• docs: https://nvidia.github.io/cuda-quantum/latest/using/basics/kernel_intro.html :

  What’s the difference between a quantum kernel and a quantum circuit? Every quantum circuit is a kernel, but not every quantum kernel is a circuit.

• docs: https://nvidia.github.io/cuda-quantum/latest/using/python.html
• pypi: https://pypi.org/project/cudaq/

NVIDIA/cuda-q-academic: https://github.com/NVIDIA/cuda-q-academic :

• https://github.com/NVIDIA/cuda-q-academic/blob/main/quick-start-to-quantum/01_quick_start_to_quantum.ipynb

Sources that may be useful:

• https://github.com/tequilahub/tequila/tree/master/src/tequila/simulators/ simulator_cudaq.py ?
  • docs: https://nvidia.github.io/cuda-quantum/latest/api/default_ops.html
• https://github.com/tequilahub/tequila/tree/master/src/tequila/ml
  • docs: https://nvidia.github.io/cuda-quantum/latest/using/integration/cuda_gpu.html
• https://github.com/tequilahub/tequila/tree/master/src/tequila/optimizers
  • docs: https://nvidia.github.io/cuda-quantum/latest/examples/python/optimizers_gradients.html
  • docs: https://nvidia.github.io/cuda-quantum/latest/api/languages/python_api.html#optimizers

[kottmanj]

Hi, the Cudaq support is currently being developed. We hope it will be ready soon, but it will potentially still take a little time. Should be there at some point in December. I flag it here once online.

But if you want to try it yourself, you need to create a new backend in src/tequila/simulators derived from simulator_base.py. I would just check out how the qulacs backend is integrated.