Unique memristor design with analog switching shows promise for high-efficiency neuromorphic computing

Researchers at Wuhan University have developed advanced memristors featuring analog switching and high on/off ratios, utilizing two-dimensional van der Waals metallic materials as cathodes. These memristors employ silver as the top anode and indium phosphorus sulfide as the switching medium. This design introduces a high diffusion barrier that modulates silver ion movement, enabling analog switching with on/off ratios up to 10⁸, over 8-bit conductance states, and attojoule-level power consumption. Chip-level simulations of convolutional neural networks demonstrated high image recognition accuracies, highlighting the potential of these memristors in neuromorphic computing applications.

Nature

  • Related Posts

    AWS’s Generative AI Expansion

    Amazon Web Services (AWS) is expanding its generative AI capabilities to compete with Microsoft and Google. At its annual re:Invent conference, AWS announced third-generation Trainium processors for training large language models and introduced AI tools designed to automate complex tasks,…

    Amazon’s AI Supercomputer Initiative

    Amazon is collaborating with Anthropic to develop Project Rainer, one of the world’s largest AI supercomputers. This system utilizes Amazon’s Trainium 2 AI chips and aims to be five times more powerful than current models used by Anthropic. Amazon has…

    You Missed

    AI Agent Startups Trends 2023–2026

    AI Agent Startups Trends 2023–2026

    The Rise of Generative UI Frameworks in 2025–26

    The Rise of Generative UI Frameworks in 2025–26

    Will OpenAI Prism accelerate scientific research?

    Will OpenAI Prism accelerate scientific research?

    Considering AI and Communism

    Considering AI and Communism

    Order in the Age of AI

    Order in the Age of AI

    Where Should AI Memory Live?

    Where Should AI Memory Live?

    2026 Will Be the First Year of Enterprise AI

    2026 Will Be the First Year of Enterprise AI

    Does the Age of Local LLMs Democratize AI?

    Does the Age of Local LLMs Democratize AI?

    Data Science and Buddhism: The Ugly Duckling Theorem and the Middle Way

    Data Science and Buddhism: The Ugly Duckling Theorem and the Middle Way

    Google’s Gemini 3: Launch and Early Reception

    Google’s Gemini 3: Launch and Early Reception

    AI Governance in Corporate AI Utilization: Frameworks and Best Practices

    AI Governance in Corporate AI Utilization: Frameworks and Best Practices

    AI Mentor and the Problem of Free Will

    AI Mentor and the Problem of Free Will

    The AI Bubble Collapse Is Not the The End — It Is the Beginning of Selection

    The AI Bubble Collapse Is Not the The End — It Is the Beginning of Selection

    Notable AI News Roundup: ChatGPT Atlas, Company Knowledge, Claude Code Web, Pet Cameo, Copilot 12 Features, NTT Tsuzumi 2 and 22 More Developments

    Notable AI News Roundup: ChatGPT Atlas, Company Knowledge, Claude Code Web, Pet Cameo, Copilot 12 Features, NTT Tsuzumi 2 and 22 More Developments

    KJ Method Resurfaces in AI Workslop Problem

    KJ Method Resurfaces in AI Workslop Problem

    AI Work Slop and the Productivity Paradox in Business

    AI Work Slop and the Productivity Paradox in Business

    OpenAI’s “Sora 2” and its impact on Japanese anime and video game copyrights

    OpenAI’s “Sora 2” and its impact on Japanese anime and video game copyrights

    Claude Sonnet 4.5: Technical Evolution and Practical Applications of Next-Generation AI

    Claude Sonnet 4.5: Technical Evolution and Practical Applications of Next-Generation AI

    Global AI Development Summary — September 2025

    Global AI Development Summary — September 2025

    Comparison : GPT-5-Codex V.S. Claude Code

    Comparison : GPT-5-Codex V.S. Claude Code

    【HRM】How a Tiny Hierarchical Reasoning Model Outperformed GPT-Scale Systems: A Clear Explanation of the Hierarchical Reasoning Model

    【HRM】How a Tiny Hierarchical Reasoning Model Outperformed GPT-Scale Systems: A Clear Explanation of the Hierarchical Reasoning Model

    GPT‑5‑Codex: OpenAI’s Agentic Coding Model

    GPT‑5‑Codex: OpenAI’s Agentic Coding Model

    AI Adoption Slowdown: Data Analysis and Implications

    AI Adoption Slowdown: Data Analysis and Implications

    Grokking in Large Language Models: Concepts, Models, and Applications

    Grokking in Large Language Models: Concepts, Models, and Applications

    AI Development — August 2025

    AI Development — August 2025