IBM has been a world leader in the field of quantum technology for years and they have developed various solutions to prove their placement as an industry leader. Recently, the move in quantum technology has been achieving quantum utility and large-scale commercialization for quantum computers. As of last year, IBM was able to run circuits beyond exact classical simulations.

Full-stack software for quantum technology at IBM

Early this year, the first version of the Qiskit SDK 1.x was released as an open-source development kit that allowed people to build quantum circuits and create simulations for complex computations. Today, over 600,000 users build with Qiskit, and over 2,900 research papers have been published based on findings from build processes.

Right now, IBM is taking the bold step of making Qiskit a high-performance tool and serving as the full-stack software across all quantum systems. Users can now use Qiskit embedded with middleware software to run all the quantum computing services while having access to new generative AI code assistance tools.IBM believes that Qiskit will drive its vision to bring useful quantum computing to the world.

Since its introduction in 2017, Qiskit has served various research organizations and individual researchers as a pathway to learning more about quantum computing. It has now become the preferred open-source quantum computing software according to the Open Source Software Survey. However, quantum computing is maturing and Qiskit is now being applied in quantum science for scientific discoveries. Now, with IBM’s over 100 qubit quantum computers, Qiskit can be used to get real value from quantum computers while experts can apply it to their domains for scientific discovery.

The process for making Qiskit the full-tack software for quantum computing begins when problems are mapped to quantum circuits using the Qiskit SDK 1.x. After that, optimizing the circuits with performance improvements to the Qiskit SDK and the new Qiskit Transpiler Service is done. This process makes the system faster. For example, binding and transpiling are now 39x faster than it was two years ago with Qiskit SDK 0.33. The next process is to execute workloads using the Qiskit Runtime Service which executes primitive queries with advanced error mitigation built-in.

In 2023, Qiskit Serverless was introduced which will provide the adequate resources required to run quantum circuits. Qiskit must continue on its path to maturity to enable it to perform optimally for production workloads as users reuse their code. In the coming year, the building blocks of Qiskit patterns will become Qiskit Functions such that they are components on the serverless layer that users can access as part of their workflows. Qiskit Functions will be provided by IBM and other third-party integrations to solve certain computational problems which may include custom error-handling routines, post-processing techniques, and many more.

To further support developers and researchers, IBM has introduced the Qiskit AI Code Assistant in alpha to increase the frictionless process on Qiskit while making quantum programming easier. The Qiskit Code Assistant is trained on an LLM model based on IBM Granite and powered by watsonx. The training is done using Qiskit SDK 1.x and IBM quantum features with integration with both Visual Studio Code and Jupyter Lab.

IBM continues to develop great solutions in quantum computing and promises to reveal more in the coming months to improve Qiskit to become the highest-performing quantum software globally. Eventually, the vision is to see that the IBM Quantum Platform combining Qiskit and IBM Quantum features equals utility-scale workflows.