The U.S. Defense Advanced Research Projects Agency, or DARPA, is taking a new path in quantum tech. The agency is funding 19 teams under its Heterogeneous Architectures for Quantum (HARQ) program. The goal is to move past the idea that one type of qubit can power the full system.
Claim 30% Off TipRanks
Forget margin or options. Here's how the pros trade IBMInstead, DARPA wants to integrate different qubit types into a single system. This mirrors how modern chips work today, where CPUs, GPUs, and custom chips each handle a task. As DARPA Program Manager Justin Cohen said, “Qubit technologies each have their own distinct advantages, but no single approach can deliver everything needed for large-scale, high-performance quantum systems.”
This shift could have a wide impact on firms like Alphabet (GOOGL), International Business Machines (IBM), and IonQ (IONQ), all of which focus on different quantum paths.
A Shift Toward Hybrid Quantum Systems
So far, most quantum firms have focused on one qubit type. Each method has clear strengths, but also key limits. For example, some qubits are fast but lose data fast, while others are more stable but slower.
Now, DARPA is pushing for a system that uses each qubit for what it does best. In this model, one part of the system may handle speed, while another handles accuracy. Over time, this could help move quantum tech from lab tests to real use in fields like drug design, energy, and defense.
At the center of this plan is a software effort called MOSAIC. This workstream will build tools that decide how to split tasks across different qubit types. The aim is to create systems that can run better than any single tech alone.
Over the next 24 months, teams will test how these mixed systems can scale. If this works, it may open the door to more useful quantum apps.
Hardware Links Remain a Key Challenge
At the same time, DARPA is working on the hardware side through its Quantum Shared Backbone effort. This part of the program focuses on linking different qubit systems.
This is not a simple task. Each qubit type operates differently, and moving data between them can lead to loss or errors. As a result, the program is focused on building high-quality links that keep data stable as it moves across the system.
Groups from Harvard University, the University of California, Berkeley, and other top schools are part of this effort. Their work will focus on new materials and methods to keep signals clean and stable.
In simple terms, the success of this program will depend on how well these links perform. If the system can move data with high accuracy, then the full vision of a mixed quantum system may become real.
What This Means for Investors
For investors, this shift suggests the quantum race may not be about a single winner. Instead, it may favor firms that can work within a shared system.
Companies like IonQ may benefit if their tech fits into a broader setup. At the same time, large firms such as Alphabet and International Business Machines may benefit from their strengths in both hardware and software.
In the long term, this approach could reshape how value is created in quantum tech. Rather than a single platform, the market may move toward a full-stack system in which hardware, software, and links all play a role. For now, DARPA’s move signals a clear change in how the field is thinking about scale and real use.
We used TipRanks’ Comparison Tool to follow key names in the space, such as Xanadu Technologies (XNDU), Rigetti Computing (RGTI), and D-Wave Quantum (QBTS). The group helps track how the field is moving toward system design, AI tools, and real market steps.
