Insider Brief
- Qinara has introduced Qronos, a deep reinforcement learning–based quantum circuit optimiser that significantly reduces circuit size across all tested benchmarks.
- Qronos outperformed Qiskit, TKET, Quartz, and Quarl, achieving 73–89% gate-count reductions and producing circuits 42–71% smaller than alternatives.
- The tool is hardware- and gate-agnostic, enabling quantum circuit compression for NISQ-era devices, with full benchmark data publicly released on GitHub.
PRESS RELEASE — Qinara (Qinara Ltd, UK) is pleased to announce a new milestone in quantum circuit optimisation. The company’s quantum circuit optimiser, Qronos™, is a proprietary deep reinforcement learning-based tool for reducing the size of quantum circuits.
Universal improvements in quantum circuit compression
Qronos significantly outperformed alternative optimisers that were tested, including Qiskit, TKET, Quartz, and Quarl, for all quantum circuits. Across the quantum circuits that were optimised, Qronos reduced gate count in the range from 73% to 89%. Similarly, the quantum circuits optimised by Qronos were from 42% to 71% smaller than any alternative. On average, quantum circuits optimised by Qronos were around half the size compared to the alternative methods.
Qinara has shared its results in full, including QASM files for all pre- and post-optimised quantum circuits. All circuits and their corresponding SVGs can be found here: https://github.com/Qinara-Inc/Qronos-Benchmark
The challenge being addressed in the NISQ era
In the Noisy Intermediate-Scale Quantum (NISQ) era, quantum computers are not error free. This means it is not yet possible to successfully execute large and complex quantum circuits. However, Qronos can be used to optimise (or ‘compress’) larger quantum circuits down to a size that permits effective execution on today’s quantum computers.
Hardware and Gate Agnostic
Qronos is both hardware and gate agnostic. Therefore, Qronos can be used to optimise quantum circuits for execution on any gate-based quantum computing stack and for any quantum gate set (including a universal gate set).
We invite colleagues in academia and industry to get in touch, should they wish to use Qronos. The team would be pleased to arrange access and to receive feedback.
Qinara: Hybrid Classical-Quantum Computing Software
Qinara, based in Edinburgh, develops quantum computing software tools and applications, specialising in hybrid classical-quantum solutions. The company is redefining how quantum technology can complement and enhance existing infrastructure.
Qronos has been developed by Swapnil Deshmukh and Rahat Santosh. The propriety technology behind Qronos is patent pending.
The wider team would like to thank those in academia and industry who have generously provided thoughts and feedback during the development of Qronos. Thanks is also due to Edinburgh Innovations, and the company’s investors, including Old College Capital.
Expressions of interest are welcome from organisations, including those developing a quantum computing stack or a cloud-based quantum platform, who would like to learn more about Qronos.
