I wanted to look at the matrix representation of CNOT gate as defined in Qiskit.

from qiskit import Aer
from qiskit.circuit import Gate
from math import pi
qc = QuantumCircuit(2)
c = 0
t = 1

q_0: ──■──
q_1: ┤ X ├

import qiskit.quantum_info as qi

op = qi.Operator(qc)

Operator([[1.+0.j, 0.+0.j, 0.+0.j, 0.+0.j],
          [0.+0.j, 0.+0.j, 0.+0.j, 1.+0.j],
          [0.+0.j, 0.+0.j, 1.+0.j, 0.+0.j],
          [0.+0.j, 1.+0.j, 0.+0.j, 0.+0.j]],
         input_dims=(2, 2), output_dims=(2, 2))

I am a bit confused, as I expected to see \begin{matrix} 1 & 0 & 0 & 0\\ 0 & 1 & 0 & 0\\ 0 & 0 & 0 & 1\\ 0 & 0 & 1 & 0 \end{matrix}

What Qiskit output as the matrix representation of CNOT looks to me like CNOT with the first qubit as target and second qubit as control.


2 Answers 2


It is because the state ordering convention. In qiskit, the states are ordered as $|00\rangle$, $|01\rangle$, $|10\rangle$, $|11\rangle$.

If you flip the values of c and t in your code, you will get the matrix you were expecting.


Qiskit uses Little Endian convention, in which the n-th qubit is written on the left side of the tensor product and the zero qubit is written on the right side: $$ q_{(n-1)}⊗…⊗q_1⊗q_0 $$ Thus, having three qubits such that the first qubit is in state 1, and the second and third qubits are in state 0, we will write this in Qiskit as $|001⟩$.

As for the $CNOT$ matrix, assuming we follow the Little endian convention, $q_0$ is the control qubit and $q_1$ is the target qubit, the truth table will look like this:

enter image description here

There is a nice rule how to construct controlled gates in Qiskit. Assuming that we have a one-qubit gate $U$ represented by matrix: $$U=\left(\begin{array}{cc} u_{00}&u_{01}\\ u_{10}&u_{11} \end{array}\right)$$ then its controlled version in Qiskit will have following matrix: $$C_u=\left(\begin{array}{cccc} 1&0&0&0\\ 0&u_{00}&0&u_{01}\\ 0&0&1&0\\ 0&u_{10}&0&u_{11} \end{array}\right)$$ So if $NOT$ gate has following matrix: $$U=\left(\begin{array}{cc} 0&1\\ 1&0 \end{array}\right)$$ Then $CNOT$ will be: $$C_u=\left(\begin{array}{cccc} 1&0&0&0\\ 0&0&0&1\\ 0&0&1&0\\ 0&1&0&0 \end{array}\right)$$


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.