DecaQ.ai reports achieving a 2.045-second median full-pipeline time for a complex quantum workload, encompassing workload generation, execution, and verification across five runs. This result utilized 108 logical target qubits within the company’s DecaQuasar BB3 architecture to perform quantum phase estimation derived from the FeMoco enzyme, a key component in biological nitrogen fixation. The QPE-4M workload incorporated 54 non-Clifford T gates within its 270 state-preparation gates, demonstrating the system’s capacity for advanced quantum operations; the company notes the benchmark records no dense global statevector allocation.
DecaQuasar Achieves 2.045-Second Median with 108-Qubit FeMoco QPE
DecaQ reports a median full-pipeline time of 2.045 seconds. The workload’s design incorporated superposition, entanglement, and non-Clifford gate preparation, pushing the boundaries of current quantum processing capabilities. The QPE-4M workload prepared a state with support across 2⁵⁴ computational basis components, representing 54 electrons in each. Eight entangled pair links connected the target-node partition, with a Schmidt rank of 256, demonstrating substantial quantum coherence within the system, FeMoco says.
The Hamiltonian used in the calculation comprised 162 generator groups, expanding to 270 Pauli components, including 216 anticommuting pairs, and spanned two target nodes alongside a QPE control node. During 80 precision rounds, the execution record detailed 12,960 generator-group evaluations and 21,600 expanded-component evaluations, highlighting the scale of the computation. The execution record reports this suggests an efficient memory management strategy within the DecaQuasar architecture, avoiding a common bottleneck in quantum computation.
The team determined an eigenphase of 27/524288, returned as an 80-bit binary string, demonstrating the system’s ability to produce precise numerical results. DecaQ clarifies that this result concerns phase evaluation for the specific FeMoco-derived workload and does not represent a complete ground-state energy calculation for the entire FeMoco molecule, according to the company. The company states in accompanying documentation that the OpenQASM attachment specifies target-state preparation, while the report separately describes the QPE-4M Hamiltonian execution.
See today’s quantum computing news on Quantum Zeitgeist for the latest breakthroughs in qubits, hardware, algorithms, and industry deals.
