Record Quantum Compilation

Parity Twine Optimizer

Record efficiency, straight out of the box. Parity Twine Optimizer is a Quantum Optimization Application Function built for both dense and sparse problems. Its hardware-native Parity Twine compiler eliminates SWAP overhead entirely, delivering best-known gate counts and circuit depth on any hardware connectivity, no manual tuning required. Now easy to try as a qiskit function on IBM Quantum. Fewer gates, less depth, on any hardware connectivity.

IBM Qiskit Function

Parity Twine Optimizer process pipeline

Optimization Problem

Pre-Processing

Parity Twine Compilation

Product details

QPU Run

Solution

Solve your optimization problem

Record compilation

Run larger problems on real hardware

Highest efficiency

The Parity Twine Optimizer minimizes circuit depth and two-qubit gate overhead, enabling larger optimization problems to run efficiently on today’s quantum processors. More of your quantum resources are spent solving the problem and not swapping qubits.

Higher performance, lower compilation overhead

Record optimization

Built on the state-of-the-art Parity Twine technique, the optimizer reduces gate count and circuit depth while preserving algorithm quality. The result is faster execution, lower error accumulation, and better performance for quantum optimization workloads, while consuming less classical compilation time.

Adaptive and hardware independent

Optimized for any connectivity

The Parity Twine Optimizer supports arbitrary problem graph connectivities and adapts to your target hardware topology. Whether your device is a grid, ladder, hexagonal, or custom architecture, it generates connectivity-aware quantum circuits and optimizes the compilation strategy dependent on the problem graph density.

Try it on IBM Quantum

Start your 30 days test phase now

The basis: ParityQC Architecture

The ParityQC Architecture, the generalized version of the LHZ architecture, provides a fundamentally new way to solve real-world problems on quantum computers, defining both the hardware layout and the algorithms. It is a quantum computing architecture that represents a new pathway to work efficiently with every currently available quantum computing platform, from NISQ towards fully error-corrected devices in the future.

More about the ParityQC Architecture

Parity Twine Optimizer via IBM Qiskit Function


# 1. Load the Parity Twine Optimizer function
catalog = QiskitFunctionsCatalog(
channel="ibm_quantum_platform",
instance="QISKIT_IBM_INSTANCE",
token="QISKIT_IBM_TOKEN",
)
function = catalog.load("parityqc/parity-twine-optimizer")

# 2. Set up the run configuraiton
backend_name = "ibm_fez"
problem = {"()": 3, "(0,)": 1, "(0, 1)": 2, "(1, 2)": -1}
variable_type="spin"
# Default options
options = {}

# 3. Run the Qiskit application function
function_job = function.run(
problem=problem,
backend_name=backend_name,
variable_type=variable_type,
options=options,
)

# 4. Retrieve the results
result = function_job.result()

For additional info click here to see the tutorials.

[1.0.2] – 2026-09-15
# Notes
This release marks the first stable production version of the Qiskit application function Parity
Twine Optimizer. Future releases expected to introduce

  •  higher order problem compilation,
  •  additional optimization strategies,

# Features

  • Solve sparse and dense QUBO problems with the efficient Parity Twine Compiler

# Documentation

  • Getting started user guide.
  • Tutorial with benchmark.

@software{parity_twine_optimizer,
author = {{Parity Quantum Copmuting GmbH}},
title = {Parity Twine Optimizer},
year = {2026},
version = {x.x.x},
url = {https://parityqc.com},
note = {{Software for connectivity-aware quantum circuit
optimization based on the Parity Twine framework.}}
}

Parity Twine Optimizer FAQ

ParityQC Set New Record Benchmark using IBM Quantum Computer with the Largest Quantum Fourier Transform ever Reported. Quantum computing is rapidly transitioning from pioneering work in the lab to the era of industrial development. This new benchmark shows that growth not only continues but accelerates.

Read the paper

Register for the IBM Flex or Premium Plan and then choose the Parity Twine Optimizer within your IBM function catalog.

You can test the Parity Twine Optimizer for 30 days for free via the IBM Platform if you have a Flex or Premium Plan. You can request a free trial here.

For IBM Flex & Premium Plan users the Parity Twine Optimizer will be available directly via the IBM Platform. When you exceeded your free test phase feel free to contact us via Email to find your personal product solution.

No additional license is required, if the Parity Twine Optimizer is called via the IBM Quantum platform during the 30 days free trial. Reach out to us via Email, if you plan heavy usage and/or integration of the Parity Twine Optimizer in your own stack.

With an IBM Platform Flex or Premium Plan you can request the free trial here. If you don’t have an IBM membership we are happy to help you out with further information via Email.

The Parity Twine Optimizer is only available for IBM Flex or Premium Plan users with a valid Account ID. If you are interested in the Parity Twine Technology but don’t have a valid IBM account ID, reach out to us directly.

Locate your Account ID by logging into your IBM Cloud account and going to: https://cloud.ibm.com/account/settings. The Account ID is a 32-character unique identifier.

Qiskit Functions can currently only be used with an active Premium or Flex plan. If you have questions or need more information on ParityQC products, feel free to contact us at any time.

Your 30 days test phase is over and you want to integrate the Parity Twine Optimizer into your daily workflow? We offer several license options for companies. Feel free to contact us via Email we find a solution for every (quantum) problem.

Following your free trial

If you have exceeded the 30-day free trial and would like to integrate the Twine Compiler into your daily workflow, we offer several license options for companies. Please feel free to contact us, we can solve every (quantum) problem.