Munich-based startup hzero exited stealth mode on November 9, 2026, announcing plans to develop software that manages quantum processors and high-performance computing (HPC) environments in an integrated way. The company aims to coordinate jobs and computing resources between quantum computers and conventional machines in order to improve performance across the whole system.
Integration of quantum computing and HPC is already beginning to reach the practical stage. In May 2026, quantum computer maker IQM announced an integration service that lets a quantum computer be treated as a node in the job scheduler Slurm. Against this backdrop, hzero is looking to differentiate itself not by simply connecting quantum computers to existing computing infrastructure, but by optimizing job allocation and resource management after the connection is made.
Optimizing operations after quantum and HPC are connected
On hzero's website, the company explains that connecting quantum computers to conventional computing resources is only the starting point for integration. Even when each individual resource performs well, a hybrid system that combines several resources may not deliver its full performance. To address this, the company is developing a mechanism that coordinates job execution and resource allocation through middleware and a shared runtime, operating the whole system in an integrated manner.
Resource allocation decisions draw on three kinds of information: application requirements, the real-time state of computing resources, and goals set by the operator. Operators can prioritize goals such as increasing the number of completed jobs, improving execution quality, and raising resource utilization. Because these goals can involve trade-offs, the design appears intended to adjust allocation to the operator's policy. The company also plans integration with existing job schedulers, application runtimes, and quantum software stacks.
The company is led by CEO Bruno G. Taketani and CTO Martin Rüfenacht. The management team has experience in quantum hardware and control technology, HPC, and system software development. Its backers include three firms: LEA Ventures, FORWARD.one, and Leo Capital. LEA Ventures presents hzero as a pre-seed investment and says it is the firm's first investment in the quantum field. hzero's website also states that the company receives support from the European Union (EU).
hzero is currently calling for collaboration from both quantum computer makers and HPC operators. It is seeking product integration, joint development, and technology partnerships with manufacturers, and technical evaluation, integration with existing infrastructure, and participation in pilot deployments from HPC centers and operators.
No concrete data yet to back up performance improvements
In the sources reviewed — hzero's website, reporting by The Quantum Insider, and LEA Ventures' announcement — no funding amount, specific customers, or figures showing performance improvements have been disclosed. What is currently known includes the management team's backgrounds, the investors, the call for pilot deployments, and the basic design approach of the software.
hzero names the performance the whole system delivers to users, and the economic value that results from it, as the yardstick for judging its technology. Completed jobs, execution quality, and resource efficiency would serve as metrics for quantitatively evaluating the effect. However, none of the three sources reviewed presents concrete measurement results based on these metrics.
For example, it is unclear by what percentage the number of completed jobs would rise compared with existing operating methods, how much execution quality would improve, or under what computing workloads and hardware configurations the evaluation was carried out. Judging the effect of new coordination software requires comparison with conventional operating methods using the same computing resources and workload conditions. Even if performance data emerge from pilot deployments, the comparison baseline and measurement conditions will be key factors in evaluating them.
The three sources above are the ones reviewed, and other interviews, investor materials, and similar documents may contain additional information. The fact that no performance data have been published while pilot participants are still being recruited does not in itself indicate a technical problem. Still, the publicly available information alone does not allow an assessment of how much the company's software could improve on existing operating methods.
As Slurm integration advances, what does hzero add?
A key point in evaluating hzero's business is how it differs from existing technologies that integrate quantum computers and HPC. On May 12, 2026, IQM announced an integration service that allows its Radiance quantum computer to be used as a Slurm node. With this mechanism, a quantum computer can be managed by the job scheduler in the same way as CPUs and GPUs. It is built on QDMI (Quantum Device Management Interface), an interface for handling quantum devices from software, and according to IQM it is already in production at the Leibniz Supercomputing Centre (LRZ).
The Munich Quantum Software Stack (MQSS), which includes QDMI, is also advancing the standardization of job submission and device control for gate-based quantum computers. According to official materials, software using QDMI can query the characteristics and constraints of a quantum device and adjust its behavior as they change. For HPC centers whose needs are met by the existing Slurm and MQSS, the case for adopting new management software is weaker.
hzero, on the other hand, aims not only to connect multiple computing resources but also to adjust job allocation according to their operating state at any given moment and the operator's goals. Standardizing connection and job submission, and allocating resources with whole-system performance in mind, are technically different challenges. However, no independent material has been found showing that integration through existing Slurm and MQSS cannot achieve equivalent coordination.
For this reason, hzero's technology is likely to be especially useful in environments that combine several types of quantum computers with conventional computing resources and need to switch operating goals depending on the use case. In a setup that merely manages a single vendor's quantum computer through Slurm, existing mechanisms may be sufficient.
In Japan as well, environments linking quantum computers and supercomputers are being built out. RIKEN began operating the GPU supercomputer "ROQUO" on June 19, 2026, and on September 3 announced the start of operations for QunaSys's "QURI SDK Enterprise." As a result, an environment now exists in which the supercomputer Fugaku and ROQUO's compute nodes can connect to quantum computers. However, these announcements do not address the challenge of optimizing operations across multiple computing resources, and hzero has not disclosed concrete plans for the Japanese market.
One benchmark for evaluating hzero's technology will be the results of pilot deployments, expected to be published by around November 2027, roughly a year after the company left stealth mode. If an HPC center shows concrete figures for improvements in completed jobs, execution quality, and resource efficiency compared with existing operating methods, it would be possible to judge the practical value of the company's software. In particular, how much it reduces the burden on operators of individually tuning resource allocation when different types of quantum computers are added will be an important indicator of its benefit.