Classiq launches a Fault Tolerance Engine to plan quantum programs for error-corrected hardware

Started by LivMorgan_Face, Today at 07:11 PM

Previous topic - Next topic

0 Members and 2 Guests are viewing this topic.

Topic: Classiq launches a Fault Tolerance Engine to plan quantum programs for error-corrected hardware   Views(Read 23 times)
Active members in this topic:
LivMorgan_Face(1)

LivMorgan_Face

Quantum software company Classiq has announced what it calls a Fault Tolerance Engine. The idea is to take a quantum program written at a high level and work out exactly what it would need to run on future error corrected hardware. That sounds dry, but it addresses one of the biggest practical questions in the field. How big a machine do you need to run something useful?

The engine converts programs from Classiq's modelling environment into concrete fault tolerant execution plans. It decides how logical qubits are laid out, how interactions are routed and how operations are scheduled. Then it calculates the physical cost, including the number of physical qubits required, error correction cycles, code distance, runtime, accumulated error rates and the number of T-gates and magic states needed. In other words, it turns a neat diagram into a detailed construction plan. Engineers will appreciate that

Classiq says the result is an architecture aware plan rather than a rough estimate based on formulas. That matters because the gap between a neat algorithm on paper and what has to run on a machine is enormous once error correction is added. Chief executive Nir Minerbi said fault tolerance changes the question quantum software has to answer, because you need to understand what a program becomes when every logical operation has to be protected. That is a refreshingly honest framing for a software company

No hardware partners, prices or benchmark numbers were included in the announcement, so it is hard to judge how accurate the estimates will be in practice. But tools like this could help companies decide which problems are worth pursuing and roughly when the hardware will be ready for them. It fits with the theme of the Nature Electronics comment we discussed, about moving from engineered advantage to practical utility

Is this the kind of tool that will bring practical quantum computing closer, or mainly useful for planning and investor presentations? And does anyone here work with quantum software at all? I would be curious to hear from anyone who has tried Classiq's platform