.jpg?w=1024)
IonQ has found a faster way to get different types of quantum hardware talking.
In an experimental demonstration published last week, quantum computing hardware maker IonQ announced a hardware link that achieved an average heralded entanglement rate of 1,032 pairs per second between two fundamentally different types of qubits: a trapped barium-138 ion and a solid-state diamond quantum memory.
The result marks the first memory-assisted interconnect to break the kilohertz threshold, surpassing previous trapped-ion benchmarks by a factor of four.
How IonQ's quantum interconnect works
Building large-scale quantum computers has long run into a physical wall: cramming thousands of fragile qubits onto a single processor creates severe cross-talk and control issues. The alternative is modular quantum computing, which links smaller chips using optical fibers.
Traditional interconnects require two independent photons (one from each processor) to arrive and interfere simultaneously at a beam splitter. Because photon loss occurs on both ends, success scales quadratically with transmission efficiency, slowing connection rates to a crawl.
IonQ avoided this bottleneck by shifting to a heterogeneous approach. Trapped ions excel at processing and maintaining coherence, while silicon-vacancy (SiV) color centers in diamond nanophotonic cavities excel at interacting with light.
Instead of waiting on two matching photons, a single 493.5-nanometer photon from the barium ion undergoes quantum frequency conversion to 737.45 nanometers, converts to early and late time-bins, and reflects off the SiV memory cavity. Detecting that single reflected photon confirms entanglement, shifting success scaling from quadratic to linear.
“Moving qubits through photons will be necessary in any large-scale quantum computer, and this demonstration sets the foundation for some amazing work to come,” said Chris Monroe, Chief Scientist at IonQ and Gilhuly Family Presidential Distinguished Professor at Duke University.
What eWEEK found: Speed alone won't make quantum networks scalable
IonQ's demonstration marks progress toward connecting different types of quantum hardware, but faster entanglement generation alone does not guarantee a scalable quantum network.
The company reported an average of 1,032 heralded entangled pairs per second, with a target-state fidelity of 87.9%. Heralding means a detection event signals successful entanglement, while fidelity measures how closely the resulting quantum state matches the intended state.
The results highlight a key challenge in distributed quantum computing: generating entanglement quickly while maintaining the quality needed for reliable operations. Faster connections could reduce delays between quantum processors, but imperfect states introduce errors that must be managed.
Performance measure | Reported result | Significance |
| Entanglement rate | 1,032 pairs per second | Demonstrates the speed of heralded entanglement generation |
| Target-state fidelity | 87.9% | Shows the quality of the generated states, with errors remaining |
| Hardware connection | Trapped barium ion and diamond quantum memory | Demonstrates entanglement between different quantum hardware platforms |
| Network scale | Experimental link | Larger, multi-node operation remains to be demonstrated |
The ability to connect different qubit technologies is particularly significant. Quantum hardware platforms have different strengths, and a reliable interconnect could eventually allow researchers to combine specialized processors and memories within a larger system.
Rather than relying on a single qubit architecture for every function, future quantum computers could use different technologies for computing, storage, and communication.
Scaling the approach will require researchers to maintain performance across additional nodes while managing photon loss, synchronization, and hardware complexity. Diamond-based quantum memories can also require specialized cryogenic cooling, adding another engineering consideration.
For enterprise technology leaders, the achievement is best viewed as progress in the hardware needed for distributed quantum computing, not evidence of deployment readiness. The next milestones will be improvements in fidelity, reliable multi-node connections, and demonstrations that maintain performance as systems expand.
What comes next for IonQ's quantum interconnect
IonQ is advancing the architecture under DARPA's Heterogeneous Architectures for Quantum (HARQ) program, which seeks to develop interconnects capable of linking different quantum computing platforms, including trapped ions, neutral atoms, and superconducting systems.
“Our technology can interface with almost any qubit type, so we see this unlocking a wide range of opportunities across multiple hardware modalities, from modular computing to networked sensing and beyond,” said Mihir Bhaskar, IonQ's senior vice president and general manager of Quantum Technologies at SkyWater.
The broader goal is to move beyond individual experimental connections toward networks that can support multiple quantum devices. Achieving that will require researchers to demonstrate reliable performance across additional nodes while addressing the engineering challenges of integrating different hardware platforms.
For now, IonQ's demonstration offers evidence that heterogeneous quantum systems can establish entanglement at higher rates. The next test is whether that performance can be maintained as researchers connect more devices into increasingly complex quantum networks.
Other news: IBM has advanced to Stage C of DARPA's Quantum Benchmarking Initiative, moving its fault-tolerant quantum computing roadmap into a more rigorous phase of independent testing.