Local circuit dynamics: which simulator is fastest

Brickwork nearest-neighbour layers: lattice models, quench dynamics

8 simulators ran this circuit at 4 to 24 qubits, 16 workloads in all. Qulacs was the fastest in 6, TensorCircuit in 5 and Qiskit Aer (MPS) in 5. The tables below hold every time as it was measured, with the fastest in each row marked. Open this circuit in the interactive tool to see the hardware that could run it and what a run would cost.

Expectation value, local observable, depth 3

Time per run, by qubit count

QubitsPennyLanePennyLane LightningQiskitQiskit AerQiskit Aer (MPS)Google qsimQulacsTensorCircuit
4 qubits
PennyLane
2.07 ms
PennyLane Lightning
1.48 ms
Qiskit
865 µs
Qiskit Aer
901 µs
Qiskit Aer (MPS)
942 µs
Google qsim
1.19 ms
Qulacs
32.1 µs
TensorCircuit
159 µs
8 qubits
PennyLane
3.73 ms
PennyLane Lightning
2.18 ms
Qiskit
1.80 ms
Qiskit Aer
1.40 ms
Qiskit Aer (MPS)
1.43 ms
Google qsim
2.03 ms
Qulacs
73.1 µs
TensorCircuit
249 µs
12 qubits
PennyLane
6.19 ms
PennyLane Lightning
2.88 ms
Qiskit
3.75 ms
Qiskit Aer
2.58 ms
Qiskit Aer (MPS)
1.83 ms
Google qsim
2.93 ms
Qulacs
585 µs
TensorCircuit
378 µs
16 qubits
PennyLane
30.3 ms
PennyLane Lightning
5.58 ms
Qiskit
35.7 ms
Qiskit Aer
12.9 ms
Qiskit Aer (MPS)
2.34 ms
Google qsim
4.20 ms
Qulacs
10.7 ms
TensorCircuit
2.17 ms
20 qubits
PennyLane
697 ms
PennyLane Lightning
66.6 ms
Qiskit
936 ms
Qiskit Aer
216 ms
Qiskit Aer (MPS)
2.79 ms
Google qsim
18.9 ms
Qulacs
206 ms
TensorCircuit
63.5 ms
24 qubits
PennyLane
skip
PennyLane Lightning
2.46 s
Qiskit
skip
Qiskit Aer
4.49 s
Qiskit Aer (MPS)
3.14 ms
Google qsim
555 ms
Qulacs
4.76 s
TensorCircuit
1.16 s

Expectation value, global observable (all_z), depth 3

Time per run, by qubit count

QubitsPennyLanePennyLane LightningQiskitQiskit AerQiskit Aer (MPS)Google qsimQulacsTensorCircuit
4 qubits
PennyLane
2.48 ms
PennyLane Lightning
1.70 ms
Qiskit
996 µs
Qiskit Aer
1000 µs
Qiskit Aer (MPS)
1.08 ms
Google qsim
1.28 ms
Qulacs
42.5 µs
TensorCircuit
181 µs
8 qubits
PennyLane
4.76 ms
PennyLane Lightning
2.50 ms
Qiskit
2.11 ms
Qiskit Aer
1.57 ms
Qiskit Aer (MPS)
1.80 ms
Google qsim
2.23 ms
Qulacs
119 µs
TensorCircuit
285 µs
12 qubits
PennyLane
8.07 ms
PennyLane Lightning
3.49 ms
Qiskit
4.26 ms
Qiskit Aer
3.20 ms
Qiskit Aer (MPS)
2.48 ms
Google qsim
3.12 ms
Qulacs
1.38 ms
TensorCircuit
848 µs
16 qubits
PennyLane
43.5 ms
PennyLane Lightning
9.80 ms
Qiskit
39.7 ms
Qiskit Aer
20.9 ms
Qiskit Aer (MPS)
3.39 ms
Google qsim
5.16 ms
Qulacs
26.2 ms
TensorCircuit
33.5 ms
20 qubits
PennyLane
891 ms
PennyLane Lightning
177 ms
Qiskit
947 ms
Qiskit Aer
368 ms
Qiskit Aer (MPS)
4.32 ms
Google qsim
33.4 ms
Qulacs
535 ms
TensorCircuit
735 ms
24 qubits
PennyLane
skip
PennyLane Lightning
8.20 s
Qiskit
skip
Qiskit Aer
7.41 s
Qiskit Aer (MPS)
5.35 ms
Google qsim
934 ms
Qulacs
10.7 s
TensorCircuit
16.2 s

Gradient, local observable, depth 3

Time per run, by qubit count

QubitsPennyLanePennyLane LightningQiskitQiskit AerQiskit Aer (MPS)Google qsimQulacsTensorCircuit
4 qubits
PennyLane
10.7 ms
PennyLane Lightning
3.25 ms
Qiskit
unsupported
Qiskit Aer
unsupported
Qiskit Aer (MPS)
unsupported
Google qsim
unsupported
Qulacs
117 µs
TensorCircuit
253 µs
8 qubits
PennyLane
22.3 ms
PennyLane Lightning
4.77 ms
Qiskit
unsupported
Qiskit Aer
unsupported
Qiskit Aer (MPS)
unsupported
Google qsim
unsupported
Qulacs
307 µs
TensorCircuit
426 µs
12 qubits
PennyLane
43.4 ms
PennyLane Lightning
6.94 ms
Qiskit
unsupported
Qiskit Aer
unsupported
Qiskit Aer (MPS)
unsupported
Google qsim
unsupported
Qulacs
3.00 ms
TensorCircuit
758 µs
16 qubits
PennyLane
268 ms
PennyLane Lightning
27.7 ms
Qiskit
unsupported
Qiskit Aer
unsupported
Qiskit Aer (MPS)
unsupported
Google qsim
unsupported
Qulacs
59.6 ms
TensorCircuit
5.49 ms

Which quantum computers can run it

At depth 3, 1,000 shots

Quantum computerCan run it up toBorderline up toEstimated cost at that size
Can run it up to
32 qubits
Borderline up to
–
Estimated cost at that size
$200.30–$600.90
via AWS Braket
Can run it up to
32 qubits
Borderline up to
–
Estimated cost at that size
$200.30–$600.90
via AWS Braket
Can run it up to
12 qubits
Borderline up to
–
Estimated cost at that size
$23.80–$71.40
via AWS Braket
Can run it up to
96 qubits
Borderline up to
–
Estimated cost at that size
Not priced
this vendor publishes no rate card
Can run it up to
48 qubits
Borderline up to
–
Estimated cost at that size
Not priced
this vendor publishes no rate card
Can run it up to
48 qubits
Borderline up to
–
Estimated cost at that size
Not priced
this vendor publishes no rate card
Can run it up to
4 qubits
Borderline up to
–
Estimated cost at that size
€0.031–€0.093
via Quandela Cloud
Can run it up to
–
Borderline up to
–
Estimated cost at that size
–
Can run it up to
–
Borderline up to
–
Estimated cost at that size
–
Can run it up to
48 qubits
Borderline up to
96 qubits
Estimated cost at that size
from $0.015
via IBM Quantum. 0.01s at maximum throughput, so this is a floor
Can run it up to
64 qubits
Borderline up to
156 qubits
Estimated cost at that size
from $0.421
via IBM Quantum. 0.26s at maximum throughput, so this is a floor
Can run it up to
48 qubits
Borderline up to
156 qubits
Estimated cost at that size
from $0.421
via IBM Quantum. 0.26s at maximum throughput, so this is a floor
Can run it up to
8 qubits
Borderline up to
64 qubits
Estimated cost at that size
$0.725–$2.17
via AWS Braket
Can run it up to
32 qubits
Borderline up to
48 qubits
Estimated cost at that size
$1.90–$5.70
via AWS Braket
Can run it up to
20 qubits
Borderline up to
–
Estimated cost at that size
$1.75–$5.25
via AWS Braket

Sizes are the grid's own: 4, 8, 12, 16, 20, 24, 32, 48, 64, 96, 128, 156 qubits. A device can run a circuit when a clean run is more likely than not, and is borderline when it is better than one in fifty. The cost is for one run of 1,000 shots at the largest size in the row, as a band and not a quote.

The simulators

Each one across every circuit family

How this was measured

One host, one harness

Every simulator ran the same circuits on one host, single-threaded and core-pinned, driven through one common harness, so any overhead the harness adds is charged to the simulator. Each result was checked against a reference implementation: a simulator that ran fast but returned the wrong answer is left out of the ranking. Nothing is interpolated or rescaled. A time was measured, or it is not shown.

Run 2026-08-10, 8 cores, threads=1. numpy 2.4.6, scipy 1.18.0, torch 2.13.0, numba 0.66.0, pennylane 0.45.1, pennylane_lightning 0.45.0, qiskit 2.5.1, qulacs 0.6.14, tensorcircuit 1.8.0, cudaq 0.15.1. One host and one thread make the comparison reproducible. It is not a prediction for your laptop.

The interactive tool shows the same measurements for any workload you pick, with the quantum hardware that could run it and what a run would cost.

Hardware figures each carry the source they came from and the date they were measured. The oldest in this set is 626 days old. Costs are bands, not quotes.

Host: Linux-6.8.0-1051-aws-x86_64-with-glibc2.35, 8 cores, threads=1, core-pinned. numpy 2.4.6, scipy 1.18.0, torch 2.13.0, numba 0.66.0, pennylane 0.45.1, pennylane_lightning 0.45.0. Every simulator was driven through one common harness, so any harness binding overhead is charged to the simulator rather than hidden.

© 2026 EigenOrbit. All rights reserved.