Solver engine · Preview · Selection & Portfolio
QG-SELECT.
Which subset do we pick? QG-SELECT chooses the best combination from an intractable number of options — portfolios, features, candidates — against your risk and capacity limits.
Overview
Selection & Portfolio.
Same core, newer engine
QG-SELECT shares QG-CORE's intake, formulation and backend routing with the more mature engines.
Benchmarked first
We measure against your current selection process before you change anything.
Your constraints
Risk limits, budgets, and regulatory constraints you already operate under — respected, not simplified away.
Canonical problems
What QG-SELECT solves.
Industries
Where it applies.
Methods
Quantum-inspired selection.
QG-SELECT compiles selection and portfolio problems to QUBO/Ising form — the same combinatorial shape as knapsack and subset-sum — and solves them with quantum annealing and quantum-inspired search over vast candidate spaces.
Research behind this engine
A quantum stock-portfolio application developed at QuMatrix — the working code QG-SELECT is being productized from.
Proof of value
No published number yet — and we're saying so.
QG-SELECT is in preview: the underlying selection research is mature, but we don't yet have an independently benchmarked figure to publish for it, the way we do for QG-ROUTE and QG-SCHED. The honest version is this — same archetype, same-caliber techniques, and when we run it against your data, that benchmark is the number that counts.
FAQ
Straight answers.
Is this actually quantum?
QG-SELECT is quantum-inspired and hybrid quantum-classical by design, on the same QG-CORE orchestration as every other engine. We're transparent about which solver handled which job.
What's the maturity?
Preview. The research and formulation are ready; we're onboarding design partners to run the first benchmarked pilots before we publish a number.
Can I still run a pilot today?
Yes — a benchmark on your data is exactly how a preview engine earns its first published number. Get in touch to scope one.
What data do you need to start?
Your candidate set, the objective and constraints, and your current selection method's results. Anonymized or sampled data is fine for a first benchmark.
Be one of the first benchmarks for QG-SELECT.
Thirty minutes to scope a pilot, a clear benchmark on your own data, and an honest read on whether it's worth going further.
Contact us →