CEA and Alice & Bob target the software gap between quantum and supercomputing
The French partnership will connect cat-qubit tools with existing high-performance computing and test industrially relevant physics problems.
France’s Atomic Energy Commission and Paris quantum company Alice & Bob announced a research partnership on 17 September to integrate quantum software with high-performance computing systems. The immediate goal is not to replace a supercomputer. It is to build a workflow in which classical and quantum resources can be used together and each receives the tasks it is best placed to handle.
The partners say Alice & Bob’s software stack will be connected through Bull’s Qaptiva platform. Qaptiva is designed as a hardware-independent layer for writing and directing quantum workloads across different back ends. Reuters reports that the project also gives the French ecosystem an alternative path in a field where Nvidia’s CUDA-Q has become influential.
Why many-body physics is the test case
A second workstream will focus on many-body physics, where interactions among numerous particles become difficult to model. The CEA says the partners want to develop algorithms that could run realistically on early fault-tolerant quantum systems around 2030. Potential applications include materials discovery, chemistry and complex industrial simulation, but the announcement does not claim that a useful quantum advantage has already been demonstrated.
Alice & Bob specialises in cat qubits, an architecture designed to suppress one important class of error at the hardware level. France agreed in June to acquire an early fault-tolerant machine from the company for installation in a European computing centre in 2027. That procurement and the new research partnership are related steps, but they are not the same contract.
The near-term value is integration
The software layer matters because laboratories and companies need to connect future processors to the computing environments they already use. A hardware-agnostic orchestration system can reduce the cost of rewriting workflows for every machine. It can also make performance comparisons easier, although claims about portability will need to be tested as systems mature.
The clearest near-term result will be technical integration and published use cases, not a general-purpose quantum computer. The CEA and Alice & Bob are building the route by which researchers could submit hybrid workloads and evaluate whether a quantum processor adds value. Timelines beyond the announced 2027 installation and the research target toward 2030 remain development goals rather than guaranteed delivery dates.
Sources & context
Reporting and reference material used for this article. Context sources do not independently confirm every news claim.
- CEA: Alice & Bob partnership announcement, 17 September ↗Primary announcement defining the two research workstreams, Qaptiva integration and target use cases.
- Reuters: CEA and Alice & Bob quantum-supercomputing software, 17 September ↗Independent reporting and market context, including the software-ecosystem rationale.
- CEA: France acquires an Alice & Bob quantum computer, 17 June ↗Primary background on the separate 2027 installation and cat-qubit architecture.
- Alice & Bob: Bull collaboration, 24 June ↗Company context on Qaptiva and hybrid-computing plans; not independent corroboration of performance claims.
Written for WHIF from the linked material. This article does not claim on-the-ground reporting. Our editorial standards.