A complex task that would normally require a top-tier classical supercomputer cluster grinding away for over 100 hours—how long would it take if we switched to a quantum computer? The answer: less than 3 minutes.
Recently (May 6, 2026), quantum computing software company Q-CTRL announced that, using IBM's quantum computer, they have tangibly solved a real-world problem with commercial value, achieving an absolute speedup of up to 3,000 times and truly realizing 'Practical Quantum Advantage'.
Image | Q-CTRL achieved a speedup of up to 3,000 times (Source: Q-CTRL)
So, what's the real weight of a 3,000x speedup?
Roughly a third of the computational power of all global supercomputers is currently consumed by chemistry and materials simulations.
To develop higher-performance batteries, more efficient solar panels, or new types of power generation equipment, scientists must figure out how electrons interact inside materials.
However, this represents a black hole-like computational bottleneck on classical computers.
Because the microscopic behavior of electrons is extremely complex, if high-precision simulation is required, traditional supercomputers hit a hard ceiling at simulating about 20 electrons; any more and the calculation simply bogs down.
So for years, the materials science community has had to make do by continuously optimizing 'approximate solving' software.
Since classical computers aren't suited for this, why not let quantum computers—which follow the same microscopic physical laws—give it a shot? In theory, these problems are precisely the kind of high-efficiency solving domain where quantum computing excels (falling into the BQP complexity class).
But reality is harsh: current-stage quantum computers are extremely unstable, prone to noise and errors that cause algorithms to crash.
The traditional method for dealing with errors is 'error mitigation', but this technique significantly slows down the quantum computer's operational speed, completely squandering the speed advantage quantum computing originally offered.
This is why, for many years, the quantum computing world could only prove itself by tackling problems with no commercial value—such as Google's claim of 'quantum supremacy' in 2019. These efforts proved that quantum computers could run fast, but they couldn't solve any real-world problems.
Image | Material structure (Source: Q-CTRL)
So, how did Q-CTRL break the deadlock?
Q-CTRL this time chose a classic conundrum called 'Fermionic Simulation', widely recognized as a perfect candidate for quantum computing to show its muscle.
Q-CTRL didn't opt to brute-force through hardware deficiencies, nor did it develop new application-layer algorithms. They liken themselves to 'the VMware of the quantum world,' specializing in foundational performance management infrastructure software.
This software can proactively suppress errors during the algorithm's runtime, rather than laboriously compensating for them after they occur, thus successfully preserving the speed advantage that quantum computing so desperately craves.
This is like fitting an extremely unruly F1 race car with the most advanced ESP (Electronic Stability Program), ensuring not only that the car stays on track but also retaining its lightning-fast speed.
Image | As the 'resolution' of the classical simulation increases, the agreement between the quantum and classical simulations also increases. Increasing resolution leads to better agreement with the quantum computer and shorter runtimes for the classical simulation. It was validated that the quantum computer can run up to 3,000 times faster than the classical computer. (Source: Q-CTRL)
To prove this logic works, Q-CTRL staged a face-off against the industry's strongest opponents.
Representing the traditional camp was the software ITensor, widely acknowledged as the most top-tier classical algorithm in materials science, based on the Time-Dependent Variational Principle (TDVP).
Developed by the Flatiron Institute, this software has a stellar track record, having underpinned over 1,250 academic papers since its release in 2015.
Representing the quantum camp was an IBM quantum computer utilizing 120 qubits, equipped with Q-CTRL's error-correction software. It simulated a maximum of 60 interacting electrons and executed over 10,000 two-qubit quantum logic operations.
The outcome of the duel was crystal clear: To get the classical supercomputer to reach computational accuracy matching the quantum calculation (with an error rate within roughly 1%), the classical software's computation time exploded catastrophically, taking over 100 hours.
In contrast, the IBM quantum computer, including all data preprocessing, communication, and hardware execution time, took a mere two and a half minutes.
Image | The quantum computer output used to simulate interacting electrons surpassed any previous demonstration in scale (120 qubits), evolution time, and resolution (Source: Q-CTRL)
A 3,000x speedup, combined with meeting industry-standard accuracy, jointly proves the 'practical quantum advantage' claimed earlier.
This means enterprise clients, materials engineers, and chemists finally have an economically meaningful reason to choose quantum computing.
It's no longer a theoretical extrapolation requiring companies to spend money 'betting on the future,' but a productivity tool that can deliver real ROI for solving high-value business problems right now.
As this infrastructure software is set to be integrated as a feature into the IBM quantum platform for public use, quantum computing has finally reached the point where it's time to calculate some 'serious business'.
References:
Contact & Tips: Qtumist_info@163.com
Further Reading