Qibo, a new open source framework for quantum simulation with hardware acceleration

We are very excited to present Qibo, the quantum simulator made by some members of Quantic in collaboration with researchers from the Technology Innovation Institute in Abu Dhabi and the University of Milano. The full paper can be read in arXiv, and the source code has been uploaded to GitHub.

Qibo is a new open-source software for fast evaluation of quantum circuits and adiabatic evolution which takes full advantage of hardware accelerators. In this work we introduce a new quantum simulation framework that enables developers to delegate all complicated aspects of hardware or platform implementation to the library so they can focus on the problem and quantum algorithms at hand. This software is designed from scratch with simulation performance, code simplicity and user friendly interface as target goals. It takes advantage of hardware acceleration such as multi-threading CPU, single GPU and multi-GPU devices.

Qibo has plenty of functionalities that will make the work with quantum circuits much user-friendlier (up), as well as some predefined models (down)

Qibo is constructed as a stack with the structure below. This structure allows the software to have a high level API together with some useful algorithms serving as tools to be used by quantum developers. Then, backends and operators, constructed on top of TensorFlow, manage all the processes needed to obtain as much efficiency as possible by specializing operators for CPU and GPU. The abstraction layer performs the operations themselves.

Qibo includes all precision and hardware configurations available for other programming languages in the same package. In addition, switching from one configuration to another is easily done by writing a pair of lines of code. Every configuration is optimal for some conditions.

The performance of Qibo is comparable or superior to previous programming languages for quantum simulation. For an instance of Quantum Fourier Transform (QFT), the best results are obtained for large numbers of qubits. In the case of adiabatic evolution, also great advantages have been achieved.

The future work serving as continuation for this first release of Qibo will include a scheme to send Qibo-jobs to quantum hardware through a remote server and a scheduler. This way, many users will be able to submit their calculations to a quantum processor.