Resource Analyzer
Target to model resource estimate
Enter a target logical error rate and a loss per cycle. The analyzer returns model values: code distance, cats per logical qubit and hardware modes. Model value: abstract circuit model with assumed loss heralds. Not a hardware operating point.
Model estimate
Enter parameters and click Estimate.
About these numbers
The analyzer is a model, not a validation:
- Model: loss threshold 0.79 percent per cycle (0.898 percent uncorrected), from surface-code runs with stim and PyMatching in an abstract circuit model with assumed loss heralds. Not a hardware operating point.
- Component simulation, separate from this model: QuTiP and Strawberry Fields agree on small-cat fidelity to about 4e-5.
- Component simulation, separate from this model: master-equation (mesolve) evolution agrees with the closed-form loss channel to 1e-9.
- Model assumption: the erasure and silent-Pauli split is a Poisson closed form times correction factors from a simulated bred cat; the qLDPC option divides hardware modes by an assumed factor of 10, and the third result card counts logical qubits in an assumed budget of 50,000 hardware modes.
- A pytest regression suite in Dyber's research repo pins these model outputs; that checks the code, not the physics.
Model values: abstract circuit model with assumed loss heralds. Not a hardware operating point. At the default 0.5 percent loss per cycle the model gives p_erase 1.88 percent, p_pauli 6.0e-4, distance 53 and 5,617 cats per logical (model threshold 0.79 percent per cycle, or 0.898 percent uncorrected). See the docs for the assumptions.