What is a qubit?

A bit versus a qubit

A classical bit is always in one of two states, 0 or 1, and you can read it as often as you like without changing it.

A qubit has two basic states as well, written |0⟩ and |1⟩ (the angle brackets are standard quantum notation, called "ket" notation). The difference is that a qubit can also be in any superposition of the two:

|ψ⟩ = α|0⟩ + β|1⟩

Here α and β are the amplitudes. When you measure the qubit, you get 0 with probability |α|² and 1 with probability |β|², and those two probabilities always add up to 1.

An example

A qubit with α = β = 1/√2 is in an equal superposition. Measure it and you get 0 half the time and 1 half the time. After the measurement, the qubit stays in whichever state you got: measuring it again gives the same answer.

So far that sounds like a coin toss. The difference is that amplitudes can be negative (or, more generally, complex numbers). Two qubits can both give 50/50 results but have amplitudes with different signs, and a later operation will treat them completely differently. That sign, called the phase, is what makes interference possible, and interference is the engine of every quantum algorithm.

What a qubit is not

It is tempting to say a qubit "stores 0 and 1 at the same time", so a quantum computer holds twice as much data per qubit. That is misleading. When you measure a qubit you get one ordinary bit out, never more. A result known as Holevo's bound shows that n qubits cannot give you more than n bits of readable information.

What a qubit does give you is a richer way to process information before you read it out.

Why more qubits matter so much

To describe n qubits fully, you need 2ⁿ amplitudes, one for every possible pattern of 0s and 1s. Two qubits need 4, ten need 1,024, and 300 qubits need more amplitudes than there are atoms in the observable universe. That is why even a modest quantum computer cannot be simulated exactly on a classical one, and why entanglement between qubits matters so much.

What qubits are made of

A qubit can be any physical system with two well-defined states that can be put in superposition. The main approaches today are:

• Superconducting circuits: tiny loops of metal cooled close to absolute zero. Used by IBM and Google.
• Trapped ions: single charged atoms held by electric fields and controlled with lasers. Used by Quantinuum and IonQ.
• Neutral atoms: uncharged atoms held in arrays by beams of light called optical tweezers.
• Photons: particles of light, with information stored in their polarisation or path.
• Spin qubits: the spin of single electrons in silicon chips similar to ordinary ones.

Each approach trades off speed, error rate, how long qubits last and how easily they can be scaled up. No single winner has emerged yet.

Why qubits are so fragile

A qubit's superposition is easily disturbed by its surroundings. Any interaction that "leaks" information about its state acts like a tiny measurement and destroys the superposition. This is called decoherence, and fighting it is one of the main jobs of quantum error correction.

Picturing a qubit

A single qubit's state can be drawn as a point on a sphere, called the Bloch sphere (you can move one yourself in our free Bloch sphere simulator). The north pole is |0⟩, the south pole is |1⟩, and every other point is a superposition. It is the most useful picture to keep in your head as you learn.