Quantum Computing with Python

Python is the working language of quantum computing. Every major framework is Python-first: Qiskit, PennyLane, Cirq, and PyQuil. These are the top-rated courses for learning quantum computing through Python, ranked by rating.

Why Python is the language of quantum computing

No other language comes close to Python's dominance in the quantum computing ecosystem. IBM built Qiskit in Python. Xanadu built PennyLane in Python. Google's Cirq is Python. Rigetti's PyQuil is Python. This convergence is not accidental. Python's scientific computing stack (NumPy, SciPy, Matplotlib, Jupyter) makes it the natural environment for the linear algebra, simulation, and visualization that quantum computing requires.

Jupyter notebooks are particularly well suited to quantum learning. You can write a circuit, execute it against a simulator, plot the results, and annotate what you're seeing, all in a single document. Every major platform provides Jupyter-compatible tutorials and textbooks, so getting started requires no special setup beyond a standard Python installation.

The practical consequence for learners is that there is more Python quantum content than for any other language: more courses, more tutorials, more open-source examples. Whether you are a complete beginner or an experienced developer, Python is the right starting point.

Which Python quantum framework should you learn first?

The honest answer depends on your goals, but Qiskit is the right default for most learners. It has the most courses, the most active community, the most complete documentation, and it gives you direct access to IBM's fleet of real quantum hardware. If you do not have a specific reason to use another framework, start with Qiskit.

PennyLane is the best choice if your interest is quantum machine learning. It is built around automatic differentiation and integrates with PyTorch and TensorFlow, making it a natural fit for researchers who want to train quantum-classical hybrid models. Xanadu, the company behind PennyLane, also offers strong free learning resources through the PennyLane Codebook.

Cirq is Google's framework and is the primary tool for researchers working with Google quantum hardware or studying the algorithms that Google's team publishes. It is more research-oriented and assumes more background than Qiskit, but the documentation is thorough.

PyQuil is used in the Rigetti ecosystem. It has a smaller community than the others, and Rigetti's hardware access program has changed over time, so it is generally not the best starting point unless you have a specific reason to work with Rigetti systems.

What Python skills do you need?

Basic Python is enough to start. If you can write a function, use a loop, and work with lists and dictionaries, you have the foundation for beginner quantum courses. Most courses introduce the quantum-specific library features as they go, so you are not expected to know Qiskit or PennyLane before you begin.

New to Python? Start here first If you cannot yet write a function or a loop, learn the basics before a quantum course, it will save you a lot of friction. Two free, beginner-friendly starting points: the official Python tutorial for a structured reference, or Automate the Boring Stuff with Python, a hands-on course you can read online for free. A week or two on functions, loops, lists, and dictionaries is plenty. For the fuller picture of what to know first, see our prerequisites guide.

NumPy is helpful but not required. Quantum states are represented as vectors and matrices, and numpy is the standard tool for working with those, but most beginner courses abstract that away behind library calls. You will see NumPy in intermediate courses when you start looking at statevectors directly or implementing custom gates.

Complex numbers appear frequently in quantum mechanics, but you do not need to be fluent in complex analysis before starting. Most introductory courses treat complex amplitudes informally at first, building intuition before introducing the mathematics. If you want to go deeper into the theory, brushing up on complex numbers and basic linear algebra will pay off as you progress.

Python quantum computing courses

Courses using Qiskit, PennyLane, Cirq, or PyQuil, ranked by rating

Frequently asked questions

Do I need to know Python to learn quantum computing?
Basic Python is enough. If you can write a loop, call a function, and work with lists, you have everything required for most beginner quantum programming courses. Libraries like Qiskit and PennyLane are designed to be approachable, and most courses teach you the quantum concepts alongside the Python patterns you need.
What is the best Python framework for quantum computing?
Qiskit is the most widely taught and has the largest community, the most courses, and direct access to IBM quantum hardware. PennyLane is the best choice for quantum machine learning. Cirq suits research contexts involving Google hardware. PyQuil is used in the Rigetti ecosystem but has a smaller community.
Can I run quantum circuits on real hardware from Python?
Yes. Qiskit connects to IBM Quantum hardware for free through IBM's cloud service. PennyLane supports multiple hardware backends including IBM, Amazon Braket, and IonQ. Cirq connects to Google's quantum processors through a research access program. All of these work from a standard Python environment.
How is quantum programming different from classical Python programming?
In classical Python you write instructions that run step by step and read the exact values of variables. In quantum programming you construct circuits made of quantum gates, execute the circuit, and collect measurement results from many shots. You cannot read a qubit's state mid-circuit without collapsing it, and the results are probabilistic. The Python syntax is familiar, but the mental model is different.

Start with a hands-on example

A concrete project to run before starting a course.

Beginner · 10 min · Qiskit

Building a Quantum Random Number Generator

Use a Hadamard gate and measurement to generate truly random bits: numbers that are not computed from a seed but decided by quantum measurement. Covers single-bit QRNG, multi-byte output, rejection sampling for arbitrary ranges, and how to run on real IBM hardware.