High-performance optoelectronic computing chips: research progress, key challenges, and future trends
Modern computing systems are increasingly constrained by the mismatch among computation, communication, and energy, which has motivated the exploration of optoelectronic computing as an alternative high-performance computing paradigm. By combining the ultrahigh bandwidth and parallelism of photonics with the control, memory, and programmability of electronics, optoelectronic computing chips have shown promising potential for high-throughput and energy-efficient information processing. This review surveys recent progress in optoelectronic computing from six key perspectives: higher computing power, higher computational precision, lower power consumption, richer functional integration, greater functional reconfigurability, and emerging photonic computing paradigms. We summarize representative device and system-level strategies for implementing optical and optoelectronic computing, including advances in multiplexed computing architectures, low-loss signal processing, precision enhancement, large-scale integration, and new computational schemes beyond conventional architectures. Particular attention is given to how different design choices shape the achievable performance of optoelectronic computing chips and to the main challenges that still limit their scalability and practical deployment. By organizing recent developments along these six directions, this review aims to provide a clearer pathway of the current state of the field and to outline possible pathways toward future high-performance optoelectronic computing systems.
Authors
- Yingjie Hong
- Zhiqi Zhu (ORCID: https://orcid.org/0009-0001-6258-2507)
- Haisong Jin
- Wenrui Wu (ORCID: https://orcid.org/0000-0001-7245-4659)
- Shuai Meng (ORCID: https://orcid.org/0009-0001-0052-6008)
- Haorong Peng
- Huifu Xiao
- Mingyang Wang
- Jiajun He
- Chaoyi Li
- Yonghui Tian
- Huanmei Cao
- Xudong Zhou
Institutions
- Lanzhou University (CN)
Publication Details
- Journal
- PhotoniX Synergy
- Published
- 2026-09-13
- DOI
- https://doi.org/10.1007/s44519-026-00008-4
- Primary Topic
- Neural Networks and Reservoir Computing
- Type
- article
- Field-Weighted Citation Impact
- 0.00
Funders
- National Natural Science Foundation of China