# A versatile large-scale coherent Ising machine from microwave to visible and telecom wavelength bands

**Authors:** Haoran Zhang, Jian Li, Haotao Zhu

PMC · DOI: 10.1038/s41377-026-02225-5 · Light, Science & Applications · 2026-02-24

## TL;DR

A new optical computing platform efficiently solves complex optimization problems and works across various applications and wavelengths.

## Contribution

A versatile coherent Ising machine platform is demonstrated with stable performance across multiple domains and wavelengths.

## Key findings

- The platform achieved a 55% success rate for 100-vertex Max-Cut problems.
- It maintained stable performance for over 8 hours.
- It was validated in molecular docking and credit scoring applications.

## Abstract

A versatile coherent Ising machine platform based on degenerate optical parametric oscillators has emerged as a powerful approach for solving NP-hard combinatorial optimization problems across diverse applications. Recent work demonstrated breakthroughs using femtosecond laser pumping, achieving a 55% success rate for 100-vertex Max-Cut problems with stable performance maintained for over 8 h. This versatile platform supports fully connected problem mapping and has been successfully validated across multiple domains including molecular docking and credit scoring, substantiating the practical potential of optical approaches for large-scale optimization and paving the way for broad integration into industrial applications.

A coherent Ising machine platform demonstrates versatile applicability across NP-hard combinatorial optimization problems, supporting the feasibility of optical approaches to large-scale optimization.

## Full-text entities

- **Diseases:** CIM (MESH:D007859)
- **Chemicals:** lithium niobate (MESH:C091692), CIM (-)

## Full text

_Full body text omitted from this summary view._ Fetch the complete paper as Markdown: https://tomesphere.com/paper/PMC12932807/full.md

## Figures

2 figures with captions in the complete paper: https://tomesphere.com/paper/PMC12932807/full.md

---
Source: https://tomesphere.com/paper/PMC12932807