Qoro Quantum Opens Cloud GPU Simulation Tier to Accelerate Large

Quantum software and middleware developer Qoro Quantum has launched its cloud-based GPU simulation tier for general availability, providing managed high-performance compute infrastructure designed to run tensor-network and noisy quantum circuit simulations without manual cluster administration. Integrated directly into Qoro’s Python framework, Divi ( qoro-divi ), the managed environment allows researchers to dispatch circuit workloads from local scripts to remote NVIDIA A10G and T4 GPU clusters.

Algorithm development often encounters a classical memory wall as qubit numbers and entanglement scale beyond statevector limits (where 64 qubits require approximately 295 exabytes of RAM at double precision). Transitioning to cluster-based tensor-network methods typically forces algorithm developers to configure Slurm schedulers, containers, and CUDA drivers manually. Qoro’s platform automates this infrastructure routing through Divi, abstracting hardware provisioning behind a single backend declaration while managing resource allocation dynamically based on the workload’s entanglement profile.

In technical validation benchmarks, the team simulated a 64-qubit non-integrable transverse and longitudinal Ising chain where non-local Jordan-Wigner terms rule out free-fermion classical shortcuts. While parallel CPU nodes handled low-entanglement circuits efficiently, GPU acceleration was required to sustain higher matrix product state (MPS) bond dimensions (χ = 512) to curb late-time energy drift. Free developer accounts receive 1,000 compute credits upon onboarding to evaluate hybrid workflows.

Benchmark data and code examples are detailed in Qoro’s Technical Blog here , with developer documentation available on the Divi Documentation Portal here .