PKTron 13.0.0 Launches With Major Upgrades: CPBN v2, New Quantum Noise Models, IBM Hardware Bridge and 60-Qubit Clifford Performance

PKTron 13.0.0 has been released with a major expansion of its quantum simulation, noise modelling, benchmarking and hardware-integration capabilities, introducing CPBN v2, new quantum noise models, an IBM hardware bridge, stabilization improvements and significant performance optimizations.

The new release is described as a stabilization, noise-modelling, hardware-bridge and performance release, while preserving the features and modules introduced throughout previous PKTron releases.

According to the release documentation, v13 brings together new noise models, advanced CPBN simulation, calibration and statistical analysis, hardware comparison tools and performance-focused simulation methods.

What Is New in PKTron 13.0.0?

PKTron 13.0.0 expands across five major areas:

Stabilization and reliability
New quantum noise models
CPBN v2
Hardware bridge and IBM integration
Performance optimization

The release also adds updated documentation, a changelog and a dedicated known-limitations section.

CPBN v2 Introduces Multi-Component Noise Environments

One of the biggest additions in PKTron 13.0.0 is CPBN v2, the expanded version of PKTron’s Computationally Pumped Bath Noise subsystem.

CPBN v2 adds support for multiple environmental bath components, configurable maximum noise rates, exact trajectory sampling, Clifford Pauli-frame sampling, calibration from data, paired statistical analysis and matched-control experiments.

The new CPBNConfig.max_rate parameter allows users to cap the noise probability. According to the release documentation, setting max_rate=1 reproduces the v12 behavior.

The new CPBNComponent and CPBNMultiScaleEnvironment functionality also allows multiple baths with different memory and pump characteristics.

This gives researchers a way to construct more complex simulated environments rather than relying on a single bath configuration.

Exact Pauli Trajectory Sampling

PKTron 13.0.0 introduces run_cpbn_trajectories, which provides exact shot-sampled Pauli trajectories.

The release documentation specifies that the bath is outcome-independent in this implementation.

This provides another route for studying CPBN behavior through individual sampled trajectories rather than only aggregate simulation results.

60-Qubit Clifford Pauli-Frame Sampling

Another major performance-oriented addition is run_cpbn_clifford.

The new Clifford pathway supports Pauli-frame sampling for circuits containing:

H, S, Sdg, X, Y, Z, ID, CX, CZ and SWAP.

The development benchmark included in the release material demonstrates the Clifford Pauli-frame backend operating at:

60 qubits, 60 gates and 1,000 shots in 0.020 seconds

The developers note that these are development-sandbox measurements on a single CPU and that performance will vary between systems.

New Quantum Noise Models

PKTron 13.0.0 introduces a new noise-modelling stack through pktron.noise_v13.

The release adds:

Thermal relaxation using T1 and T2
Coherent over-rotation
ZZ crosstalk
Correlated readout error
Correlated readout mitigation
Pauli twirling
Average gate fidelity
CPTP checking
Composable NoiseStack

These components can be combined to create more detailed noise configurations.

For example, the new NoiseStack can combine thermal relaxation, coherent over-rotation, ZZ crosstalk, depolarizing noise and correlated readout errors within one simulation.

Correlated Readout Error and Mitigation

The new CorrelatedReadout model allows users to configure both individual readout errors and pair-dependent readout effects.

PKTron also introduces:

CorrelatedReadout.mitigate(counts)

This functionality applies the configured readout model to measured counts for mitigation.

The release therefore adds both the ability to simulate correlated measurement errors and a corresponding mitigation pathway.

CPTP Validation and Gate-Fidelity Analysis

PKTron 13.0.0 also adds tools for evaluating quantum noise channels.

The new functionality includes:

CPTP validation
Average gate fidelity
Pauli twirling probabilities

This allows users to inspect characteristics of noise channels instead of simply applying them to a circuit.

CPBN Calibration and Statistical Analysis

The v13 release goes beyond noise generation by adding tools for calibration and statistical analysis.

The cpbn_calibration module provides functionality for simulating data series and fitting CPBN-related parameters.

PKTron also introduces paired statistical analysis through paired_report.

The release documentation specifies that paired tests use signed differences and a sign-flip permutation test.

A matched-control experiment is also available through matched_control_experiment.

These tools are intended to help researchers compare controlled circuit histories and determine whether a configured CPBN environment produces a measurable difference.

Importantly, the documentation notes that a matched-control result can legitimately be zero for a particular configuration or history pair.

New Hardware Bridge Connects PKTron With Qiskit and IBM Backends

PKTron 13.0.0 introduces a dedicated hardware-bridge subsystem through pktron.hardware_bridge.

It includes:

Qiskit circuit export
IBM backend runner
Matched-pair construction
Matched-pair analysis
Count comparison
Comparison reports
Safe JSON result export

The new to_qiskit(circuit) function provides a pathway for exporting PKTron circuits into Qiskit.

The ibm_runner(backend) functionality is designed to execute circuits on an IBM backend.

The release specifically addresses qubit-order handling during comparison.

The hardware bridge also provides tools for constructing matched experiments and generating structured comparison results.

Performance Improvements Target CPBN Overhead

Performance is one of the central themes of v13.

The release introduces a combined Pauli-channel fast path intended to significantly reduce CPBN overhead.

The development documentation reports an improvement from approximately 26× overhead to approximately 2× in the relevant workload.

Additional performance work includes:

Tensor-form channel appliers
A dedicated stabilizer sampler
Combined Pauli-channel execution
A new noise benchmark suite

The release notes state that the combined Pauli-channel implementation is equivalent to the v12 pathway to within 1e-10.

Development Benchmark Results

The supplied release material includes the following development-sandbox measurements:

Backend	Qubits	Gates	Shots	No Noise	CPBN
Density Matrix	8	48	–	0.075 s	0.159 s
Trajectories	12	54	256	–	0.290 s
Clifford Pauli Frame	60	60	1,000	–	0.020 s

The developers explicitly note that these measurements were taken on a single CPU in a development environment and that users should expect different results on their own hardware.

Major Stabilization Improvements

PKTron 13.0.0 also addresses several previously failing areas.

The stabilization work covers:

MPS
Surface Code
Color Code
QASM2
QPY
Molecule
VRP

The MPSSimulator SVD helper has been converted into a proper static method.

SurfaceCode now includes n_data and decode().

ColorCode now accepts a distance and an error array.

The Molecule implementation has also been rewritten around atomic numbers and n_electrons.

QASM2 Security and Parsing Improvements

One of the notable code-quality changes is the removal of direct eval() usage from QASM parameter parsing.

PKTron 13.0.0 instead routes parameters through safe_eval_param, using an AST whitelist.

The permitted expression system includes numbers, pi, addition, subtraction, multiplication, division and exponentiation, with length and exponent limits.

The release also introduces QASM2Codec.

For QPY files, invalid magic bytes now result in a ValueError.

Improved Surface-Code Reporting

PKTron also changes the way SurfaceCodeDistance.logical_error_rate is reported.

The method is now labelled:

heuristic_mc_analytical_blend

The documentation explicitly identifies this as a heuristic rather than a full decoder simulation.

That distinction is important when using the result in research or benchmarking.

Known Limitations Are Explicitly Documented

PKTron 13.0.0 includes a dedicated known-limitations section.

Several APIs remain compatibility placeholders and now emit PlaceholderWarning.

These include:

StateTomography.reconstruct
QSPAngleFinder.find_angles
LinearCombinationBlockEncoding.encode
FaultTolerantCircuit.compile
ElectronicStructureProblem.hamiltonian
ParityMapper.map
BravyiKitaev.map
ActiveSpaceTransformer.transform
FreezeCoreTransformer.transform
Z2Symmetries.taper

The documentation explicitly states that their output should not be used for research results.

CPBN Is Not Being Claimed as a Demonstrated Hardware Effect

PKTron also makes an important distinction between simulation architecture and experimental evidence.

The release documentation states that an earlier IBM hardware experiment produced a null result for CPBN history effects.

Therefore, CPBN is presented as a simulator architecture, not as a demonstrated physical effect on quantum hardware.

The project also states that noise models improve simulation realism only when their parameters match the characteristics of the device being modelled.

PKTron makes no claim that its noise models produce better results than other simulators.

Release Testing and Package Validation

The supplied PKTron 13.0.0 Colab build includes an automated build, test and publishing workflow.

The workflow:

Extracts the complete PKTron source tree.
Inspects the source.
Verifies version numbers.
Installs build dependencies.
Runs the test suite.
Builds the wheel and source distribution.
Checks package metadata with Twine.
Performs a clean wheel installation.
Runs smoke tests.
Publishes the package to PyPI.
Performs a post-publication installation check.

The build output records successful package creation, metadata validation, installation and smoke testing of the v13 functionality.

The published package is available on PyPI as PKTron 13.0.0.

PKTron 13.0.0 on PyPI

What Makes PKTron 13.0.0 Significant?

The main change in PKTron 13.0.0 is not a single isolated feature.

The release brings together several layers of quantum simulation research:

Circuit simulation → advanced noise modelling → CPBN environments → calibration → statistical analysis → matched controls → Qiskit export → hardware comparison → benchmarking

With CPBN v2, the new noise stack, hardware bridge and performance improvements, PKTron 13.0.0 significantly expands the scope of the simulator.

At the same time, the release documentation takes care to distinguish simulated behavior from experimentally demonstrated hardware effects and identifies APIs that should not yet be used for research conclusions.

PKTron 13.0.0 Upgrade Summary

Version: 13.0.0

Major upgrades include:

CPBN v2
Multi-component CPBN environments
Configurable maximum noise rates
Exact Pauli trajectory sampling
Clifford Pauli-frame sampling
60-qubit development benchmark
Thermal relaxation noise
Coherent over-rotation
ZZ crosstalk
Correlated readout error
Readout mitigation
Pauli twirling
Average gate fidelity
CPTP checking
Composable NoiseStack
CPBN calibration
Paired statistical analysis
Matched-control experiments
Qiskit export
IBM backend bridge
Matched-pair hardware analysis
Safe JSON results
Combined Pauli-channel fast path
Tensor-form channel appliers
Dedicated stabilizer sampler
Expanded benchmark suite
QASM2 safety improvements
QPY validation
MPS, SurfaceCode, ColorCode and Molecule stabilization