TL;DR
Qblox announced an integration connecting NVIDIA DGX Spark to its quantum system controller through NVIDIA NVQLink. The company reports measured round-trip latency below 7.2 microseconds on DGX Spark and below 4.0 microseconds on an RTX PRO 6000 Blackwell; independent verification and detailed test conditions were not provided in the supplied report.
Qblox announced an integration connecting NVIDIA’s DGX Spark system to its quantum system controller through NVQLink, reporting measured round-trip latency below 7.2 microseconds on DGX Spark. The company says the link can bring GPU computing into the real-time control loop used by quantum hardware, including for calibration and error-correction workloads.
Get business pricing on monitors, keyboards and dev gear
- Business-only prices and quantity discounts
- Tax-exempt purchasing
- Multiple users, one account, clear invoices
Qblox also reported round-trip latency below 4.0 microseconds using an NVIDIA RTX PRO 6000 Blackwell GPU. These figures are company-reported measurements; the announcement, as summarized by The Quantum Insider, does not provide a full test methodology, workload description or independent benchmark. The two figures refer to different hardware configurations and should not be read as a direct comparison of equivalent systems.
The integration combines Qblox’s Q1 sequence processors and LINQ with NVIDIA GPU computing, using CUDA-Q to support development on desktop hardware. Qblox says GPU-accelerated calibration routines can update qubit parameters while experiments are running, rather than waiting until they finish. It also describes sending quantum error-correction syndrome data to GPU-based decoders with low latency, with the goal of identifying and correcting errors in real time.
The companies frame DGX Spark as a desktop development platform for workloads that combine quantum computing, AI and simulation. Qblox says developers can carry the same code to GPU servers as systems scale. The announcement describes this as a development path; it does not establish that every workload will transfer unchanged or perform identically on larger systems.
Connecting GPUs to Quantum Control
The announcement addresses a practical challenge in operating quantum processors: their control systems must coordinate experiments and process information quickly, while classical computing resources handle tasks such as calibration and error decoding. Putting a GPU in that loop could let some computations run closer to the quantum hardware’s operating timescale, rather than being handled only between experiment runs.
That matters to researchers exploring real-time calibration and quantum error correction, both of which are relevant to efforts to build more reliable quantum systems. However, the reported latency figures alone do not show that the integration improves qubit performance, increases the number of logical qubits, or delivers a specific error-correction outcome. Those results depend on system configuration and workload testing that the announcement does not detail.
DGX Spark’s role is to offer a desktop route for developing and testing hybrid workloads, according to NVIDIA and Qblox. If the same software can be used on larger GPU systems as hardware grows, that could reduce friction between early experiments and later deployments. The extent of that portability remains to be demonstrated for particular applications.
Products Worth Considering
Bose QuietComfort Ultra Earbuds (2nd Gen), Wireless Noise-Cancelling Earbuds with Immersive Audio and Wireless Charging Case, Black
Logitech MX Mechanical Wireless Illuminated Keyboard, Clicky Switches
Samsung Odyssey G3 G30D 27-Inch FHD Gaming Monitor
As an affiliate, we earn on qualifying purchases.
From NVQLink Plans to Hardware
The development follows earlier Qblox work with NVIDIA. The companies announced a collaboration involving NVQLink in October 2025, according to the supplied report, and Qblox added support for NVIDIA’s CUDA-Q real-time API in March 2026. Qblox presents the new integration as a move from architecture and software support to measured performance on hardware.
NVQLink is described in the announcement as the connection between GPU computing and quantum control. Qblox’s control products manage sequences and hardware interactions, while CUDA-Q supports programming across quantum and classical computing resources. The October announcement’s specific contribution is the reported integration with DGX Spark and the latency measurements, rather than a claim that a complete fault-tolerant quantum system is now available.
“Quantum computers will scale only as fast as the classical control systems around them.”
— Niels Bultink, Qblox CEO and co-founder
Benchmark Details Still Missing
The supplied report does not specify the test conditions behind the latency measurements, including the precise workload, system configuration, measurement method or number of trials. It also does not say whether the results have been independently reproduced. The figures should therefore be treated as Qblox-reported measurements, not as a general performance guarantee for all quantum systems or tasks.
It is also unclear which researchers or customers can access the integration now, whether it is generally available, and what hardware or software requirements apply. The announcement describes potential uses in calibration and error correction but does not provide application results showing improved calibration time, error rates or logical-qubit performance.
From Demonstration to Workloads
The next evidence to watch for is detailed testing of the integration on named quantum systems and specific workloads. Published benchmark methods, independently reproduced latency measurements, and results for calibration or decoding would help clarify how the reported figures translate into research use.
Qblox and NVIDIA say developers can use CUDA-Q on desktop hardware and then scale code to GPU servers. The companies have not provided a timetable for broader availability or described a particular deployment schedule in the supplied material. Further information on access, supported configurations and demonstrations will show how readily researchers can use the system beyond the announced integration.
Key Questions
What did Qblox announce?
Qblox said it integrated NVIDIA DGX Spark with its quantum system controller using NVQLink, connecting GPU computing with quantum control.
What latency did Qblox report?
Qblox reported round-trip latency below 7.2 microseconds on DGX Spark and below 4.0 microseconds on an RTX PRO 6000 Blackwell. The supplied report does not include full test conditions or independent verification.
What could the integration be used for?
Qblox says it is intended to support hybrid quantum-classical tasks such as GPU-accelerated calibration and processing data for quantum error correction. The announcement does not report measured improvements in error rates or qubit performance.
Does the announcement mean fault-tolerant quantum computing is available?
No. The announcement describes a connection between GPU computing and quantum control, and identifies error correction as a possible workload. It does not claim that a complete fault-tolerant quantum computer is available.
Source: rss
