Cudaq Guide

CUDA-Q Onboarding Reference

Use this reference for /cudaq-guide install, test-program, gpu-sim, qpu, applications, and parallelize.

Documentation Map

Install

Instructions:

• Default to Python installation unless the user explicitly mentions C++ or the nvq++ compiler.

• After installation, always guide the user through validation with the Bell state example and confirm output shows roughly { 00:~500 11:~500 }.

• Default to GPU-accelerated targets (nvidia) unless the user is on macOS/Apple Silicon, mentions no GPU, or explicitly asks for CPU-only simulation. In those cases use qpp-cpu.

• Do not suggest cloud options unless the user has no local environment, asks about cloud access, or needs classroom/workshop access for students.

Platform notes:

• Linux x86_64 or ARM64: full GPU support with pip install cudaq plus CUDA Toolkit.

• macOS ARM64/Apple Silicon: CPU simulation only with pip install cudaq; no CUDA Toolkit needed.

• Windows: use WSL, then follow Linux instructions.

• C++ without sudo:

  Copy

  bash install_cuda_quantum*.$(uname -m) --accept -- --installpath $HOME/.cudaq

• Brev cloud workspace: log in at the NVIDIA Application Hub, open a CUDA-Q workspace, then SSH in with the Brev CLI:

  Copy

  brev open ${WORKSPACE_NAME}

  CUDA-Q and the CUDA Toolkit are pre-installed.

Classroom / academic access:

When the user is teaching a course or running a workshop and students need access without local setup, recommend:

• Brev academic workspaces: same NVIDIA Application Hub / Brev CLI flow as above, with CUDA-Q pre-installed.

• CUDA-Q Academic (https://github.com/NVIDIA/cuda-q-academic): ready-made course notebooks runnable locally or in the cloud via qBraid, CoCalc, or Google Colab (see docs/sphinx/using/quick_start.rst).

• Amazon Braket notebook instances: managed cloud notebooks with CUDA-Q available, useful when the class also targets Braket QPUs.

Test Program

Key concepts to explain:

• @cudaq.kernel / __qpu__ marks a quantum kernel compiled to Quake MLIR.

• cudaq.qvector(N) allocates N qubits in |0>.

• cudaq.sample() runs a measured kernel and returns a bitstring histogram (SampleResult).

• cudaq.run() runs a kernel with a classical return value shots_count times and returns the list of those return values.

• cudaq.observe() computes expectation value <H> for a spin operator.

• cudaq.get_state() returns the full statevector on simulators.

Kernel restrictions:

• Only a restricted Python subset is valid inside a kernel; it compiles to Quake MLIR, not normal Python.

• NumPy and SciPy cannot be used inside a kernel. Use them outside kernels for classical pre/post-processing.

• Kernels can call other kernels, but the callee must also be a CUDA-Q kernel.

For compiler internals (inspect module -> ast_bridge.py -> Quake MLIR -> QIR -> JIT), route to /cudaq-compiler if that skill is available.

GPU Simulation

To recommend the best simulation backend, consult the latest CUDA-Q simulator backend table and detailed backend sections: https://nvidia.github.io/cuda-quantum/latest/using/backends/simulators.html.

QPU

Do not dump all providers at once. Use a two-step dialogue.

Step 1: ask which technology they want:

Which QPU technology are you targeting?
  1. Ion trap         (IonQ, Quantinuum)
  2. Superconducting  (IQM, OQC, Anyon, TII)
  3. Neutral atom     (QuEra, Infleqtion, Pasqal)
  4. Cloud / multi-platform (AWS Braket, Scaleway)

Step 2: once they pick a technology, ask which provider, then use the latest CUDA-Q provider listings and examples:

• Hardware provider index: https://nvidia.github.io/cuda-quantum/latest/using/backends/hardware.html

• Cloud backend index: https://nvidia.github.io/cuda-quantum/latest/using/backends/cloud.html

• Provider examples: https://nvidia.github.io/cuda-quantum/latest/using/examples/hardware_providers.html

Then walk through that provider's setup steps.

After walking through the provider steps, always close with:

• Test locally first with emulate=True before submitting to hardware.

• Use cudaq.sample_async() / cudaq.observe_async() for non-blocking submission.

• Handle provider credentials securely: export them as environment variables in the shell session, or use a local profile or secrets manager that is not committed. Never paste tokens into shared files, logs, notebooks, or commits.

Applications

CUDA-Q ships with ready-to-run application notebooks.

Parallelize

CUDA-Q supports two distinct multi-GPU parallelization strategies. Pick based on what the user is trying to scale.

Circuit batching with mqpu

The mqpu option maps one virtual QPU to each GPU. Dispatch circuits asynchronously with qpu_id to all GPUs simultaneously.

import cudaq

cudaq.set_target("nvidia", option="mqpu")
n_qpus = cudaq.get_platform().num_qpus()

futures = [
    cudaq.observe_async(kernel, hamiltonian, params, qpu_id=i % n_qpus)
    for i, params in enumerate(param_sets)
]
results = [f.get().expectation() for f in futures]

Hamiltonian batching

For a single kernel with a large Hamiltonian, add execution= to cudaq.observe; no other code change is needed.

# Single node, multiple GPUs
result = cudaq.observe(kernel, hamiltonian, *args,
                       execution=cudaq.parallel.thread)

# Multi-node via MPI
result = cudaq.observe(kernel, hamiltonian, *args,
                       execution=cudaq.parallel.mpi)

See the CUDA-Q docs for complete working examples of both patterns.

Invocation Examples

• /cudaq-guide: print the onboarding menu and ask which topic to explore.

• /cudaq-guide install: walk through installation, defaulting to Python pip install cudaq, then validate with the Bell state example.

• /cudaq-guide test-program: build and run a Bell state kernel and confirm output shows roughly { 00:~500 11:~500 }.

• /cudaq-guide gpu-sim: recommend a simulation backend, such as nvidia for one GPU or nvidia --target-option mgpu for circuits larger than one GPU's memory.

• /cudaq-guide qpu: start the two-step QPU dialogue and read the matching hardware doc.

• /cudaq-guide parallelize: choose between mgpu for pooled memory and mqpu for many independent circuits.

• /cudaq-guide author: route to references/authoring.md.

Platform Troubleshooting

• Import error after pip install cudaq: ensure Python 3.10+ and a supported OS, Linux or macOS.

• No GPU detected: verify CUDA Toolkit and nvidia-smi; fall back to qpp-cpu.

• Multi-GPU mgpu fails: verify MPI availability and CUDA-Q target support.

• QPU submission fails: confirm credentials are set as environment variables or through the provider mechanism documented in local CUDA-Q docs.