Develop Quantum Circuits with Cirq

Cirq is a Python package for writing, manipulating, and running quantum circuits on quantum computers and simulators, built for today's noisy intermediate-scale quantum (NISQ) hardware where hardware details matter to getting good results. It provides flexible and custom gate definitions, parameterized circuits with symbolic variables, circuit transformation/compilation/optimization, hardware device modeling, noise modeling, multiple built-in circuit simulators, integration with qsim for high-performance simulation, and interoperability with NumPy and SciPy.

Use it when building, transforming, or simulating quantum circuits in Python, especially when NISQ-era hardware specifics (gate sets, noise, device topology) need to be modeled explicitly rather than abstracted away. It supports Linux, macOS, Windows, and Google Colab, requiring Python 3.11 or later. For fault-tolerant quantum computing algorithm research specifically, Google's broader Quantum AI stack offers complementary tools (e.g. Qualtran) rather than Cirq itself covering that need alone.

Install via pip install cirq (or pip install --upgrade cirq~=1.0.dev for the pre-release). A minimal example: pick a qubit with cirq.GridQubit(0, 0), build a cirq.Circuit containing a gate (e.g. cirq.X(qubit)**0.5, square-root-of-NOT) and a cirq.measure operation, then run it with cirq.Simulator().run(circuit, repetitions=20) to get measurement results back as a bitstring. Circuits print as ASCII diagrams (e.g. (0, 0): ---X^0.5---M('m')---).

Cirq is the base layer of a broader Google Quantum AI open-source stack, with tools chosen by interest area: Qualtran for quantum algorithms and fault-tolerant quantum computing (FTQC), qsim for large circuits and heavy simulation workloads, Stim for circuits with thousands of qubits and millions of Clifford operations or for quantum error correction (QEC), the OpenFermion family (OpenFermion, OpenFermion-FQE, OpenFermion-PySCF, OpenFermion-Psi4) for chemistry and materials science, TensorFlow Quantum for quantum machine learning, and ReCirq for real experiments run on Cirq. Documentation spans video tutorials, Jupyter-notebook tutorials runnable in Colab with no install, text-based tutorials on the Cirq site, a GitHub examples subdirectory, and a dedicated Experiments page. Separate reference docs exist for the current stable release (matching pip install cirq) and the pre-release build (matching pip install --upgrade cirq~=1.0.dev), and a GitHub releases page tracks the changelog for each version.

Cirq has drawn contributions from over 200 people, releases roughly every quarter, and is citable via a Zenodo-archived DOI for reproducibility in published research. It is explicitly not an officially supported Google product.

Researchers and developers writing, simulating, or running quantum circuits in Python - particularly for NISQ-era hardware where noise and device-specific gate behavior need to be modeled directly rather than assumed away.