Desks live on estimates: an option's fair value, tomorrow's tail risk, the best portfolio the mandate allows. Quantum amplitude estimation attacks exactly this class of problem — and we show you what it delivers today, on programs you can run before we ever meet.
Monte Carlo burns compute to shave error. Amplitude estimation converges quadratically faster in theory — the real question is what survives on today's machines, and we answer it with runs, not slides.
Value-at-Risk is a regulatory number. Our demonstration reads VaR(95%) straight out of a quantum distribution load — and lands on the exact classical answer, so you can check every step.
Cardinality limits, sector caps, lot sizes — the constraints are what make portfolio selection hard. As a QUBO, the constraint wall becomes part of the energy landscape itself.
We take one instrument or book you actually price — not a textbook payoff — and formulate it: distribution loading, payoff rotation, the estimator, the constraint set.
The formulation runs on simulators first, then on real hardware where shot prices are public. Every number arrives with its estimation error separated from its discretisation error.
You get a written answer to the only question that matters: at what scale, if any, does this beat the Monte Carlo stack you already run — pursue, park, or drop.
These programs are published in our algorithm library. The numbers below come from recorded executions we can reproduce on demand.
Prices a European call with quantum amplitude estimation.
Measured: The estimate separated estimation error (0.037) from discretisation error — the two are usually blurred together in vendor decks.
Reads 95% Value-at-Risk from a loaded loss distribution.
Measured: Landed on the exact classical answer, reproducibly.
Selects a constrained portfolio formulated as a QUBO.
Measured: Constraint penalties tuned until every sampled optimum was feasible.
Results are from the library items' own recorded runs on our simulator — the same one your browser uses.
Amplitude estimation's quadratic speedup is real mathematics, but today's machines cap the circuit depth it needs. At current scale, your Monte Carlo farm is still faster on production books.
Send a line about what you price and at what scale — an engineer replies, not a sales script.