Ready to Get Started?
Install Cirq with pip and start building quantum circuits in Python.
Overview
Cirq is an open-source Python library for writing, manipulating and optimizing quantum circuits, released under the Apache 2.0 license. It targets today's noisy intermediate-scale quantum (NISQ) computers, where hardware details matter: device definitions model hardware constraints, and circuits are built from Moments that you can slice and transform.
Cirq ships with wave function and density matrix simulators that can model noise, and works with Google's qsim simulator. Its Quantum Virtual Machine mocks the constraints and noise of real hardware. Google uses Cirq to run experiments on its own quantum processors.
For real hardware, the Cirq site lists AQT, IonQ, Pasqal and Microsoft Azure Quantum. Related projects include OpenFermion, TensorFlow Quantum, qsim, Stim, Qualtran and ReCirq.
Key Features
Device-aware circuits
Device definitions capture the constraints of specific hardware, so circuits respect what the device can run.
Moments and transformations
Circuits are organized into Moments, with tools to slice, transform and optimize them.
Built-in simulators
Wave function and density matrix simulators, with noise via Monte Carlo or full density matrix simulation; works with qsim.
Quantum Virtual Machine
Uses the simulators to mimic the constraints and noise of real quantum hardware.
Third-party hardware
Run on AQT, IonQ, Pasqal or through Microsoft Azure Quantum.
Algorithm library
Textbook quantum algorithms and tutorials, including a QAOA max-cut example.
Pros & Cons
Advantages
- Free and open source under Apache 2.0
- Fine-grained control over hardware details for NISQ research
- Built-in noisy simulation and a Quantum Virtual Machine
- Runs on Linux, macOS, Windows and Google Colab
- Works with qsim, Stim, OpenFermion and TensorFlow Quantum
Disadvantages
- No hosted hardware service of its own; you need an account with a provider
- The Cirq site does not describe public access to Google's own processors
- Requires Python 3.11 or later
- The repository describes it as not an officially supported Google product
- Lower-level than application frameworks; expect to write circuits by hand
Pricing
Cirq itself costs nothing:
Cirq library
Free and open source (Apache 2.0). Install with pip install cirq.
Hardware runs
Billed by the hardware provider you use, such as IonQ, AQT, Pasqal or Azure Quantum.
Details from quantumai.google/cirq and the Cirq GitHub repository, October 2026.
Best Use Cases
Google Cirq Excels For:
- NISQ algorithm research where hardware constraints matter
- Simulating noisy circuits before running them on hardware
- Quantum chemistry with OpenFermion
- Teaching circuit-level quantum programming in Python
May Not Be Ideal For:
- Users looking for a managed cloud service with hardware included
- Quantum machine learning projects that need built-in automatic differentiation
- Beginners who want high-level, ready-made applications
Comparison
Google Cirq vs Competitors
Where Cirq stands out
- Designed around device constraints and Moments for precise NISQ control
- Tight links with Google tools like qsim, Stim and OpenFermion
- Fully free and open source
Where alternatives fit better
- IBM Quantum (Qiskit): open-source SDK tied to IBM hardware with 10 free minutes per month
- PennyLane: differentiable programming for quantum machine learning
- Amazon Braket: managed service with hardware from several vendors
Screenshots & Interface
Explore Google Cirq's interface:
Frequently Asked Questions
Is Cirq free?
Yes. Cirq is open source under the Apache 2.0 license and installs with pip install cirq.
Which hardware can I run Cirq on?
The Cirq site lists Alpine Quantum Technologies (AQT), IonQ, Pasqal and Microsoft Azure Quantum. Hardware usage is billed by the provider.
Can Cirq simulate noise?
Yes. Its wave function and density matrix simulators handle noisy channels, and the Quantum Virtual Machine mimics the noise of real hardware.
Can I use Google's quantum computers?
Google uses Cirq for experiments on its quantum processors, but the Cirq site does not describe a public access program for them.
What do I need to install it?
Python 3.11 or later on Linux, macOS or Windows; it also runs in Google Colab.
Final Verdict
Cirq is a free, precise tool for circuit-level quantum programming, especially when hardware constraints and noise matter. It is a library, not a service, so you will need a separate hardware provider to run on real quantum computers.
