leaky

Leakage-aware stabilizer simulation built on top of stim.TableauSimulator.

Install

From PyPI:

pip install leakysim

From source:

git clone https://github.com/inmzhang/leaky.git
cd leaky
pip install .

Quick Example

import leaky
import stim

channel = leaky.LeakyPauliChannel(2)
channel.add_transition(
    leaky.LeakageStatus(status=[0, 0]),
    leaky.LeakageStatus(status=[0, 1]),
    "XI",
    1.0,
)

sim = leaky.Simulator(4, [channel], seed=1234)
circuit = stim.Circuit("""
R 0 1 2 3
X 0 2
CNOT 0 1 2 3
I[leaky<0>] 0 1 2 3
M 0 1 2 3
""")

sim.do_circuit(circuit)
print(sim.current_measurement_record().tolist())  # [0, 2, 0, 2]

Main Python API

• LeakageStatus: per-qubit leakage labels. 0 means computational space.
• LeakyPauliChannel: leakage transitions plus Pauli corrections on qubits that stay computational.
• Simulator: executes stim.Circuit objects while tracking leakage.
• ReadoutStrategy: controls how leaked measurement results are reported.
• generalized_pauli_twirling(...): converts Kraus operators into a LeakyPauliChannel.

Notes

• Use I[leaky<n>] ... in a Stim circuit to apply the nth channel bound to a simulator.
• Simulator(seed=...) only seeds that simulator instance.
• apply_leaky_channel(...) expects raw qubit targets and broadcasts in groups of channel.num_qubits.
• For ReadoutStrategy.RawLabel, leaked measurements are reported as 2, 3, ... instead of projected bits.

API Reference

The shipped stub file is the canonical Python API reference:

• src/leaky/__init__.pyi