Atom Computing, Diraq, IBM and IonQ Advance to DARPA’s Final Quantum Benchmarking Stage

DARPA Quantum Benchmarking Initiative logo

Six teams are now undergoing evaluation of whether their proposed systems can reach utility-scale operation by 2033.

DARPA has selected Atom Computing, Diraq, IBM and IonQ to advance to Stage C of its Quantum Benchmarking Initiative (QBI), moving four distinct quantum computing approaches into the program’s final evaluation stage. Announced October 7, the selections expand the group that will be undergoing government verification and validation of proposed utility-scale quantum computers.

The four companies join Microsoft and PsiQuantum, which advanced through DARPA’s earlier pilot program, Underexplored Systems for Utility-Scale Quantum Computing (US2QC). DARPA expanded that effort into QBI in 2024 to evaluate a broader range of quantum computing methods. US2QC’s final phase has the same technical goals as QBI’s Stage C, bringing all six companies into evaluation of whether their proposed systems can be constructed and operated as intended.

The newly selected teams represent different hardware architectures. Atom Computing, based in Boulder, Colorado, is developing scalable arrays of neutral atoms. Sydney-based Diraq uses silicon CMOS spin qubits. IBM, based in Yorktown Heights, New York, is pursuing modular superconducting processors, while College Park, Maryland-based IonQ is developing trapped-ion quantum computing.

Their advancement follows Stage B, a yearlong assessment of detailed research and development plans. DARPA examined the assumptions underpinning those plans, the technical risks each organization faces, proposed risk mitigation strategies, and the prototypes needed to demonstrate progress. Stage C extends that scrutiny to whether the proposed systems can be built according to their designs and operate as intended.

“Stage C is where we begin moving from plans to proof,” said QBI Managing Director Micah Stoutimore. “Our team will be closely scrutinizing the hardware and systems as they develop to determine whether the technology performs the way the models and roadmaps say it should.”

QBI provides funding as well as independent technical assessment. DARPA lists potential awards of up to $1 million for Stage A, $15 million for Stage B and $300 million for Stage C. These are program ceilings, rather than the amounts awarded to the four newly advanced companies. The October 7 announcement did not disclose individual contract values.

The initiative continues to evaluate companies at earlier stages. DARPA’s participant list, updated October 8, identifies seven active Stage B teams: Nord Quantique, Photonic, Quantinuum, Quantum Motion, QuEra Computing, Silicon Quantum Computing and Xanadu. Google Quantum AI and Quandela are listed in Stage A. DARPA cautions that the list reflects active work and may temporarily omit companies between completing one stage and receiving a subsequent contract.

Companies have entered the program on different schedules, resulting in staggered advancement. DARPA’s November 2025 Stage B announcement identified 11 selected companies and explained that further promotions would be announced after contracting was finalized. The latest selections therefore represent another step in an ongoing evaluation process.

DARPA emphasizes that QBI evaluates each approach on its own merits, with no predetermined number of successful teams. Stoutimore said the agency increasingly expects someone to achieve utility-scale quantum computing by 2033, while the successful teams and architectures remain uncertain. Further entrants and additional Stage C promotions are expected as the initiative continues.

Scroll to Top