A digital twin partnership wants to model quantum noise before it ever hits real hardware

[Neil57]

Two companies teaming up on the unglamorous problem of noise. Quantum Elements, a provider of AI powered digital twins for quantum computing developers, has signed a development agreement with Planckian, an Italian company building a novel superconducting quantum processor architecture, to support Planckian's error correction strategy. The deal has Quantum Elements building architecture specific noise models to characterize the physical noise environment inside Planckian's chips, accounting for coherence loss, leakage and operation level error sources

Why simulating noise is harder than it sounds

Quantum processors are getting more sophisticated, but they remain plagued by environmental noise, crosstalk between neighboring qubits, and control imperfections, all obstacles standing between today's hardware and genuine fault tolerance. Researchers normally study this by simulating quantum systems on classical computers, often through direct density matrix simulation that tracks a noisy quantum system's full state alongside its interaction with the environment. The problem is that the amount of information needed to represent the system explodes as qubit counts grow, quickly becoming computationally prohibitive

The workaround, and a proof of concept that already worked

Quantum Elements' Digital Twins technology lets researchers model noisy quantum circuit behavior with far lower computational resources while still preserving the dynamics needed to study error correction, correlated noise and decoder performance. This isn't just a theoretical pitch, a prior collaboration with AWS, USC and Harvard used a Quantum Monte Carlo accelerated digital twin to simulate a 97 physical qubit, distance-7 surface code syndrome extraction round on ordinary classical computing infrastructure. AWS reported that a brute force simulation of that same system would need to track an impossible number of density matrix entries, while the accelerated digital twin approach ran in about an hour on a single compute node

Why Planckian specifically needs this

Planckian's whole pitch is a chip architecture that strips out the control complexity and wiring overhead that normally prevents conventional superconducting processors from scaling. But as its CEO Michele Dallari points out, a genuinely new architecture also reshapes exactly what kinds of errors the system has to deal with, meaning Planckian needs a faithful, architecture specific picture of its own noise environment before it can meaningfully decide how to correct for it, evaluated on classical hardware well before actually trying to scale up the physical chips

Quantum Elements and Planckian Partner on Digital Twins for Superconducting Quantum Processorsthequantuminsider.com