June 22, 2026 -- Even the most sophisticated quantum hardware is only as powerful as the algorithms that run on it. Progress in quantum computing depends not only on building more advanced systems, but on enabling researchers to develop new methods that harness those capabilities. That’s why IBM has prioritized Open Access from the start—by puttingthe first quantum computer on the clouda decade ago, and through programs like therecently expanded IBM Quantum Open Plan, newClassroom Accounts, and IBM Quantum Credits.

The IBM Quantum Credits programinvites researchers to submit high-quality project proposals and earn free, direct access to IBM quantum computers, with awards based entirely on technical merit. In this post, we highlight four recent projects from Credits recipients that demonstrate how new algorithms and methods generated through the program are extending the reach of today’s quantum hardware.

The Credits program prioritizes high-impact, utility-scale research proposals driven by novel algorithms and methods. Applicants—typically tenure-track faculty or professional researchers—should demonstrate a clear, feasible plan for extracting meaningful results from real quantum hardware, with progress achievable in 5-10 hours of QPU time.

Simulating particle collisions on quantum hardware

The motivation.Physicists use particle colliders to probe the fundamental structure of matter, converting energy into new particles through high-energy collisions. Classical computers cannot simulate complex particle collisions from first principles, in part due to the difficulty of representing the quantum state of many interacting particles. Quantum computers show enormous potential to fill that gap.

The result.IBM Quantum Credits recipientsRoland Farrell(Caltech) andNikita Zemlevskiy(University of Washington)developed a new approachto quantum state preparation for collision simulations, introducing an algorithm that prepares the localized particle states, or “wavepackets,” required for particle scattering simulations using constant-depth quantum circuits. By leveraging W-state preparation techniques withmid-circuit measurement and classical feedforward, their method avoids the scaling bottlenecks of earlier approaches. In their experimental demonstration, the team used IBM quantum hardware to simulate particle collisions and observe, for the first time in a gate-based quantum simulation, the emergence of a new particle from the interaction. The work has since been featured at the IBM Quantum Developer Conference, where Zemlevskiy presented the results in akeynote presentation.

How IBM Quantum Credits enabled the work.Farrell and Zemlevskiy say Quantum Credits were critical in transforming their theoretical proposal into a working experimental demonstration. “Having the runtime beyond the basic IBM Quantum Open Plan was essential to refining our simulations and error mitigation strategies,” they said. Direct access to hardware for iterative experimentation proved especially valuable. “Running experiments on quantum hardware is more than just a demonstration,” Farrell said. “I have always gained insight about the system I am simulating from iterating through different attempts.”

Reconstructing mixed quantum states at scale

The motivation.To understand and validate quantum computers, we must reconstruct the states they produce—not just simplified “pure” states describing a system in isolation, but the “mixed” states that capture how it interacts with its environment. This task becomes prohibitively difficult as system size grows. Standard randomized measurement techniques can probe certain properties, but typically require costly measurements and are not adapted to full state reconstruction. Credits recipientBenoît Vermersch(Université Grenoble Alpes; Quobly), his former student Matteo Votto (Université Grenoble Alpes), and collaborators set out to develop a method for efficiently learning and characterizing large, noisy quantum states in real experimental settings.

The result.The team introduced a protocolthat uses randomized measurements to reconstruct quantum states as tensor networks—compressing them into efficient classical representations that enable extraction of global properties like entanglement and entropy without repeated measurements. In experiments, the team reconstructed entangled states on up to 96 qubits using IBM Quantum hardware, showing the method can scale far beyond previous approaches. The learned representation also captures noise and decoherence, enabling large-scale error mitigation through more efficient classical post-processing.

How IBM Quantum Credits enabled the work.This work required hundreds of thousands of experimental runs on quantum hardware, but Vermersch says the Credits program did more than provide access. “It gives you the motivation, and the right context in terms of relevant experimental parameters, to make your theoretical proposals truly adapted to an experimental scenario,” he said. “There are always things you cannot fully anticipate or control without experimental data.” Vermersch also emphasized that applying to the Credits program was straightforward, and that support from the IBM Quantum team enabled a smooth transition from proposal to experiment.

Advancing quantum simulations of complex materials

The motivation.The most interesting quantum materials are often defined by complex many-body interactions that are extremely difficult to simulate classically. A key reason is frustration, where the geometry of a system prevents particle interactions from being simultaneously satisfied. Thekagome lattice, a two-dimensional network of corner-sharing triangles, is a canonical example. It naturally encodes antiferromagnetic interactions and gives rise to highly degenerate, strongly entangled ground states. Accurately computing the ground-state energies of these lattice models is a central challenge in physics, but preparing these states on quantum hardware usually requires circuits too deep for today’s devices.

The result.Quantum Credits recipientMuhammad Ahsan(University of Engineering and Technology, Lahore; National Center for Quantum Computing)developed a scalable version of the Variational Quantum Eigensolver (VQE) algorithmthat combines a hardware-efficient ansatz with a novel Hamiltonian calibration strategy to better match current quantum hardware capabilities. By restructuring the computation into smaller, classically optimized subproblems and recombining them on quantum processors, his method reduces circuit depth while preserving accuracy. In experimental demonstrations, he computed the ground-state energy of a 103-qubit system inaccessible to exact classical methods. The results approach leading classical approximations while revealing non-classical features of the system.

How IBM Quantum Credits enabled the work.Like his fellow Credits recipients, Ahsan emphasized the importance of hardware access as a driver of innovation, as well as the ability to exchange methods and concepts with other Credits recipients and the IBM Quantum team. “Quantum Credits provided early access to processors with lower-error-rate fractional gates—genuine game-changers,” he said. Those capabilities led to more stable experiments, while direct interaction with quantum hardware shaped his approach to algorithm design and noise mitigation, ultimately enabling successful executions of large-scale experiments on real quantum systems. Using a similar approach, his current experiments investigate classically unattainable ground-state properties of larger lattices, continuing his work under the Credits program.

Exploring fundamental physics beyond classical limits

The motivation.Classical supercomputers help physicists study the fundamental forces of nature by enabling simulations of physics theories like quantum chromodynamics (QCD), which describes how quarks and gluons interact to form larger particles, such as protons and neutrons. These classical simulations have enabled tremendous progress in understanding of fundamental physics, but some key problems—including real-time dynamics and certain phase structures—remain out of reach due to challenges like the “sign problem,” which make the underlying mathematics exponentially more difficult to compute.

The result.Credits recipientIndrakshi Raychowdhury(BITS Pilani, Goa Campus; Centre for Research in Quantum Information and Technology) has spent decades developing alternative formulations of fundamental physics theories that are better suited for quantum computation.In recent work through the Credits program, she and her collaborators developed quantum simulation algorithms based on Hamiltonian formulations—using the quantum mechanical description of the physical degrees of freedom of a gauge field theory to map problems more naturally onto quantum hardware. These methods aim to preserve underlying physics while making simulations more tractable, offering a path toward studying complex quantum field theories that challenge classical methods. Importantly, the underlying methods—rooted in lattice gauge theory and Hamiltonian simulation—are broadly transferable, with potential applications ranging from QCD to quantum many-body and condensed-matter systems.

How IBM Quantum Credits enabled the work.Raychowdhury says the Credits program helped her bridge the gap between theoretical ideas and experiments. “The Quantum Credits program is amazing as it offers access to state-of-the-art quantum hardware [and tools like] Qiskit add-ons and other advanced error mitigation techniques,” she said, noting how that expanded access enabled testing across the full workflow. Her advice to other applicants: identify a clear problem where classical methods fall short, build a proof of concept, and use real quantum devices to refine and benchmark your approach.

From proposal to experiment

These examples offer both inspiration and practical guidance for researchers considering submitting an application to the Credits Program:

1. Start with a clearly defined problem
2. Develop a novel technical approach
3. Be prepared to iterate on real quantum hardware

Successful proposals require more than just a good idea. Applicants—typically university faculty and professional researchers—should first validate their approach on real quantum hardware, for example by accessing hardware for free through the IBM Quantum Open Plan. The most impactful work comes from developing new methods that can scale on real hardware.