Quantum gate
Also known as: quantum logic gate, qubit gate, quantum circuit gate
A quantum gate is an operation that changes the state of qubits, the quantum equivalent of logic gates in regular computers.
Quantum programs, called circuits, are sequences of gates followed by measurement. While the analogy to classical gates helps, quantum gates are reversible and manipulate amplitudes rather than simple 0s and 1s, and every gate on real hardware introduces a small amount of error.
Quantum gate, explained at three levels
At home
Regular computers use logic gates like AND and OR to process bits. Quantum computers use quantum gates to change qubits. A program is a series of these gates, like steps in a recipe, followed by checking the result.
At work
When companies talk about quantum computers, they often mention 'gate fidelity,' meaning how accurately gates work. Higher fidelity means fewer errors, which is essential for running useful programs.
At home
Some common quantum gates include the Hadamard gate, which puts a qubit into an equal superposition, and the CNOT gate, which flips one qubit depending on another and creates entanglement. Combining these with a few others lets you build any quantum program.
At work
Quantum hardware supports a limited 'native' gate set, and compilers translate programs into those gates. Two-qubit gates usually have higher error rates and take longer than single-qubit gates, so minimizing them is a key optimization. Circuit depth (the number of gate layers) is limited by how long qubits stay coherent.
At home
Mathematically, quantum gates are unitary matrices, which means they're reversible and preserve total probability. In my view, that reversibility is a beautiful constraint, but it also means you can't simply copy or erase quantum information the way you can with classical bits.
At work
A universal gate set, such as Clifford gates plus the T gate, can approximate any operation. Clifford-only circuits can be simulated efficiently on classical computers (the Gottesman-Knill theorem), so non-Clifford gates like T are where quantum advantage comes from, and they're expensive to implement in error-corrected systems. Resource estimates for useful algorithms are often expressed in T-gate counts.
Real-world examples
At home
Aisha tried a free online quantum programming tool. She dragged a Hadamard gate onto a qubit and measured it 1,000 times, getting roughly half 0s and half 1s, her first hands-on look at superposition.
At work
Jerome is a research engineer evaluating cloud quantum services. Instead of comparing qubit counts, he compared two-qubit gate fidelities and how deep a circuit each system could run before errors overwhelmed the results.
How it compares
| feature | Quantum gate | Classical logic gate |
|---|---|---|
| Operates on | Qubits and their amplitudes | Bits (0 or 1) |
| Reversible | Always | Often not (for example, AND) |
| Error rate | Noticeable, around 0.1% to 1% for two-qubit gates on good hardware | Negligible |
| Examples | Hadamard, CNOT, T | AND, OR, NOT |
| Copying information | Not possible (no-cloning theorem) | Easy |
Quantum gates are reversible and error-prone; classical gates are irreversible and extremely reliable.
For analysts
When comparing quantum providers, ask for gate fidelities and achievable circuit depth, not just qubit counts. I'd treat a vendor that leads with qubit numbers alone as one that may be hiding weak gate performance.
History & origin
Physicists developed the circuit model of quantum computing in the 1980s and 1990s. David Deutsch proposed quantum computational networks in 1989, and a 1995 paper by Adriano Barenco and colleagues showed that a small set of 'elementary' gates could build any quantum computation. Early experiments demonstrated individual gates in the mid-1990s: a team at NIST, led by David Wineland and Christopher Monroe, performed a quantum logic gate with trapped ions in 1995. Over the following decades, gate fidelities improved steadily, with leading systems demonstrating two-qubit gate fidelities above 99.9% by the mid-2020s. In my opinion, gate quality, not qubit count, is the most honest single measure of progress in quantum hardware.