The Tool Desk
Outbyte Driver Updater FREEScan for outdated or missing drivers - takes under a minuteDriver Scan →Outbyte PC Repair FREERepair Windows errors before they cause bigger problemsFix Now →Some links on this page are affiliate links: if you buy through them we may earn a commission, at no extra cost to you.
Diraq, the Australian quantum-computing company founded and led by silicon-qubit researcher Andrew Dzurak, is targeting its first commercial product in 2029. The company says that system should deliver genuine quantum advantage on economically meaningful workloads—not simply demonstrate a laboratory result. That is an ambitious corporate target, not a guaranteed delivery date.
Diraq’s strategy is to build dense arrays of silicon spin qubits with semiconductor manufacturing methods and integrate them with conventional data-center computing. The decisive tests will be manufacturing yield, error correction, cryogenic control, and useful performance against classical alternatives.
What Diraq’s 2029 promise actually means
“First quantum computer by 2029” can suggest that Diraq currently has no working hardware. Its later announcements are more precise: early systems are intended for data-center environments before or around that period, while the first commercial product is targeted for 2029. The target therefore appears to mean a deployable, useful system rather than Diraq’s first prototype, test chip, or laboratory demonstrator.
Free tools Windows power users keep installed
One-click scans. No signup required.
Diraq defines the goal as a machine capable of “genuine quantum advantage”—a measurable benefit on selected valuable tasks compared with the best practical classical alternative. It does not mean the machine will outperform conventional computers on every workload.
#1 Best Overall
Andrew Dzurak is Diraq’s founder and chief executive. The company says he has worked in quantum technology for more than 25 years and has led silicon-quantum research associated with the University of New South Wales, whose research and intellectual property helped form Diraq.
Diraq’s announcement describes the 2029 product target, while its NVIDIA collaboration announcement distinguishes that product from earlier systems entering data centers.
Why Diraq is betting on silicon spin qubits
A qubit is the quantum counterpart of a classical bit. It can occupy a superposition of states and be entangled with other qubits. In Diraq’s architecture, information is encoded in the spin of an electron confined in a silicon-based quantum-dot device.
Rank #2
The commercial argument is manufacturing. Silicon spin qubits are physically small and can potentially be fabricated in dense arrays using processes related to the CMOS techniques used throughout the semiconductor industry. That could offer repeatable production, access to established foundries, lower eventual unit costs, and tighter integration between quantum devices and classical control electronics.
Silicon does not remove the difficult parts of quantum engineering. A complete machine still needs high-fidelity gates, reliable measurement, initialization, error correction, cryogenic wiring and electronics, packaging, thermal management, software, and classical orchestration. Diraq’s technology roadmap describes integration with high-performance classical computers and a compact, server-rack-scale quantum-processing unit as goals.
The technical evidence—and what it does not prove
EE Times reported that Diraq and imec produced silicon quantum-dot qubits with two-qubit operation fidelity above 99%. Two-qubit fidelity measures how closely an individual operation matches its intended result. Reaching that level is an important milestone because some error-correction schemes require operation errors below particular thresholds.
It is not evidence that a complete quantum computer is 99% correct or already fault tolerant. Thresholds depend on the error-correcting code, connectivity, measurement performance, leakage, control errors, noise assumptions, and correlated errors. A small-device demonstration also does not establish million-qubit yield or system-level reliability.
Independent manufacturing context comes from a 2025 npj Quantum Information paper reporting semiconductor-based spin-qubit processors made with industrial 300-millimeter wafer processing. That work shows why foundry-compatible fabrication matters, but it does not by itself validate Diraq’s 2029 product specification. (Paper; EE Times interview coverage.)
Diraq’s stated roadmap
| Period | Company-stated objective | How to interpret it |
|---|---|---|
| 2022–2025 | Build internal capability and begin fabrication of foundry devices | Development and manufacturing preparation |
| 2025–2029 | Scale foundry manufacturing to systems exceeding 1,000 qubits and pursue early-value applications | More than 1,000 physical qubits is a hardware target, not a logical-qubit count |
| 2029 onward | Implement commercially valuable algorithms with utility-scale systems | Depends on error correction, usable runtime, and customer access |
| By 2031 | Millions of qubits | Longer-term company target |
| By 2033 | Tens of millions of qubits | Longer-term company target |
These are Diraq’s targets, not independently validated forecasts. Physical qubits are raw hardware units. Logical qubits are error-corrected units built from multiple physical qubits, and logical-qubit performance is much more relevant to useful computation. Diraq also says planned systems could support one million error-corrected operations per minute; that is a company specification or aspiration, not an independently demonstrated result.
The manufacturing and packaging bottleneck
Diraq has worked with imec and has identified GlobalFoundries as an important foundry partner. The company has cited GlobalFoundries’ 22FDX process as attractive because of its performance at cryogenic temperatures, according to EE Times.
Making many nominally similar devices on a wafer is only the first step. A scaled machine must connect and control them without exceeding a cryogenic system’s heat budget, package the chip with low-loss and low-noise connections, maintain high measurement fidelity, and keep manufacturing yield high as arrays become larger. The control electronics may need to operate close to the qubits while conventional processors coordinate error correction and applications.
Funding and commercial relationships
- Diraq announced a $20 million equity investment from Australia’s National Reconstruction Fund Corporation. The NRFC announcement confirms the investment.
- Diraq announced a letter of intent for up to $38 million in proposed U.S. CHIPS funding to develop and scale silicon-spin processors and a U.S. supply chain. This is proposed or planned support, not a completed grant. (Diraq release; NIST release).
- Diraq identifies imec, GlobalFoundries, NVIDIA, and Dell among its technology relationships. NVIDIA and Dell are described as partners for integrating quantum technology with classical data-center infrastructure, not as disclosed purchase commitments.
- Diraq says it has raised more than $100 million and holds more than 60 patents and trademarks. Those figures are company-reported. Its Series A-2 announcement is available here.
What “utility-scale” must demonstrate
For Diraq, utility-scale means economically valuable computation, not merely a large qubit number. A credible demonstration would need:
Best Value
- Stable logical qubits produced through error correction.
- Circuits deep enough to solve a defined problem.
- A repeatable result against a clearly specified classical baseline.
- Total cost and runtime that make the quantum approach worthwhile.
- Reliable customer access integrated with high-performance classical computing.
Potential application areas include chemistry, materials, energy, health, and finance, but Diraq has not publicly established which workload will first meet its advantage claim. Investors should ask what benchmark, classical comparison, error rate, and availability commitment will define success.
What could derail the 2029 target?
- High-fidelity gates on small devices may not survive at large scale.
- Wafer yield and qubit uniformity can deteriorate as arrays and interconnects grow.
- Millions of control lines, cryogenic electronics, packaging, and thermal limits may constrain density.
- Error correction may require many physical qubits for each logical qubit, reducing useful capacity.
- A laboratory speedup may not translate into an economically superior customer workload.
- Letters of intent and partnerships do not guarantee fully funded production or customer deployments.
- The company has not publicly disclosed enough detail to independently verify the complete 2029 product specification.
Milestones worth watching
- Qubit count on a single integrated chip, rather than an aggregate across separate demonstrations.
- Wafer yield and device uniformity.
- Two-qubit gate, measurement, and leakage performance at scale.
- Demonstrated logical qubits and error-correction cycle results.
- Control-electronics power consumption and cryogenic packaging density.
- A programmable system available to outside users.
- A published workload benchmark against a defined classical baseline.
- Binding customer commitments and final government funding awards.
How to experiment with quantum computing now
Diraq does not sell a consumer-accessible processor or public subscription. Readers who want hands-on access today can use third-party hardware and simulators through Amazon Braket via the AWS console. AWS describes usage-based billing, including a $0.30 per-task fee for several QPUs, device-specific per-shot charges, and simulator pricing; observed figures can change and should be checked on the current pricing page. Braket is not Diraq hardware and does not demonstrate what Diraq’s future access model or economics will be.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
Do these 3 things before closing this tab:
1Fix the driver behind crashes, sound loss and screen glitches2Clear out junk files and repair common Windows errors3Scan for outdated or missing drivers - takes under a minute
