Benchmarking quantum advantage
As claims of quantum advantage emerge, this project provides a platform-agnostic framework to collect, validate, and compare experimental results.
- Algorithmiq
- BlueQubit
- EPFL
- Flatiron Institute
- IBM
- ORNL
- RIKEN
- University of Chicago
- University of Maryland
- University of Toronto
- University of Wisconsin – Madison
- Vector Institute
- Algorithmiq
- BlueQubit
- EPFL
- Flatiron Institute
- IBM
- ORNL
- RIKEN
- University of Chicago
- University of Maryland
- University of Toronto
- University of Wisconsin – Madison
- Vector Institute
What is quantum advantage?
Quantum advantage refers to performing an information processing task more efficiently, cost-effectively, or accurately using a quantum computer than is known to be possible with classical computers alone.
But achieving this milestone requires more than raw performance - it demands trust in the output of noisy quantum devices and scientific rigor in how we validate results.
Why is it hard to verify?
Quantum advantage isn’t a single milestone - it’s a falsifiable scientific hypothesis that must be tested through rigorous experimentation. Because quantum computers tackle problems that classical systems can’t easily replicate, direct comparison is challenging. Verifying any claim of advantage therefore demands multiple points of analysis.
✏️ "The test of all knowledge is experiment" — R. P. Feynman
Three pathways to quantum advantage
To build confidence in advantage claims, this project explores three pathways for analysis. Learn more about the different paths below.
- Observable estimations 📊Trust through rigorous error control.Explore submissions that report expectation values for observables, and include mathematically grounded error bars for validating quantum computations.View the tracker
- Variational problems 🌀Certifiable quantum solutions via the variational principle.Variational algorithm submissions offer guaranteed solution bounds and enable benchmarking against classical methods - even when exact answers are unknown.View the tracker
- Classically verifiable problems 🗝️Leveraging classical resources to validate quantum outputs.Submissions in this path enable efficient validation of quantum outputs without requiring full classical simulation of the quantum process.View the tracker