Two platforms built on the same quantum-inspired core already running inside iQ Finance today: optimization for the hardest combinatorial problems, and machine learning that stays interpretable without losing accuracy.
Classical solvers like Gurobi or CPLEX handle convex, well-behaved optimization problems well. They struggle once a problem turns highly non-convex, mixes in genuinely discrete decisions, or needs variables that don't reduce cleanly to numbers.
Separately, regulated industries need machine learning models a human can actually follow, but the interpretable models that give them that clarity have traditionally sacrificed real accuracy to get there.
Native support for binary, integer, categorical, and permutation variables, for problems too discrete or too non-convex for solvers like Gurobi or CPLEX. This engine already runs in production today, inside iQ Finance.
Models a human can genuinely follow, at accuracy levels usually reserved for black-box models. One of these approaches was developed through a research collaboration with the Bank of Spain.
This isn't a hypothesis. It's already running, and already being validated.
This is the same engine that already powers iQ Finance's client-facing products today.
A public-private consortium funded by the Comunidad de Madrid, coordinated by GMV, with BBVA, Repsol, and the CSIC, applying it to logistics and route assignment.
Tested against solvers like Gurobi, not just claimed to be faster.
Logistics and routing, space mission planning, and energy, alongside interpretable ML for regulated industries.
See what's possible with a free demo
