Front-End Integrations¶

Clifft's native input is Stim-compatible circuit text with Clifft extensions for non-Clifford operations. If your circuits are already written in another quantum software framework, companion packages can translate or route supported circuits to Clifft without hand-writing that text.

These packages are maintained separately from the core clifft package and are released on their own schedule. Use their READMEs as the source of truth for the full supported operation set and current limitations.

Integration Options¶

Starting point	Package	What it provides
Stim-compatible text	`clifft`	Direct parsing, compilation, sampling, state-vector access, detectors, observables, and QEC-oriented workflows.
Qiskit `QuantumCircuit`	`clifft-qiskit`	A Qiskit `BackendV2` provider that runs supported circuits on Clifft and returns Qiskit-style results.
Cirq `cirq.Circuit`	`clifft-cirq`	A converter to Clifft circuit text plus a Cirq-style sampler facade backed by Clifft.

Use the native clifft API when you are already working with Stim-style circuit text, detector annotations, observables, post-selection, or importance sampling. Use an adapter when the circuit construction, decomposition, or surrounding workflow already lives in Qiskit or Cirq.

Qiskit¶

Install the Qiskit adapter:

pip install clifft-qiskit

Then run a supported QuantumCircuit through the Clifft backend:

from qiskit import QuantumCircuit
from clifft_qiskit import ClifftProvider

qc = QuantumCircuit(2, 2)
qc.h(0)
qc.cx(0, 1)
qc.measure([0, 1], [0, 1])

backend = ClifftProvider().get_backend("clifft")
counts = backend.run(qc, shots=1000).result().get_counts()
print(counts)

The adapter targets terminal-measurement sampling and counts. Unsupported semantics, such as mid-circuit measurement, feedforward, reset, and other non-unitary operations, are rejected explicitly.

See the clifft-qiskit repository for the current supported basis, decomposition behavior, and package-specific limitations.

Cirq¶

Install the Cirq adapter:

pip install clifft-cirq

Convert a parameter-resolved qubit circuit to Clifft text or sample it through the Cirq-style facade:

import cirq
import clifft_cirq

q0, q1 = cirq.LineQubit.range(2)
circuit = cirq.Circuit(
    cirq.H(q0),
    cirq.CNOT(q0, q1),
    cirq.measure(q0, q1, key="m"),
)

converted = clifft_cirq.to_clifft_text(circuit)
print(converted.clifft_text)

sampler = clifft_cirq.ClifftSampler(seed=123)
result = sampler.run(circuit, repetitions=1000)
print(result)

The converter supports parameter-resolved qubit circuits and common one-, two-, and three-qubit gates that map to Clifft. It does not model Cirq device, timing, calibration, qudit, arbitrary classical-control, or noise-channel semantics.

See the clifft-cirq repository for the current supported operations, conversion metadata, and package-specific limitations.