Artificial Intelligence (AI) is the domain of large resource-intensive data centres that limit access to a small community of developers^1,2. Neuromorphic hardware promises greatly improved space and energy efficiency for AI but is presently only capable of low-accuracy operations, such as inferencing in neural networks^3,4,5. Core computing tasks of signal processing, neural network training and natural language processing demand far higher computing resolution, beyond that of individual neuromorphic circuit elements^6,7,8. Here we introduce an analog molecular memristor based on a Ru-complex of an azo-aromatic ligand with 14-bit resolution. Precise kinetic control over a transition between two thermodynamically stable molecular electronic states facilitates 16,520 distinct analog conductance levels, which can be linearly and symmetrically updated or written individually in one time step, substantially simplifying the weight update procedure over existing neuromorphic platforms³. The circuit elements are unidirectional, facilitating a selector-less 64 × 64 crossbar-based dot-product engine that enables vector–matrix multiplication, including Fourier transform, in a single time step. We achieved more than 73 dB signal-to-noise-ratio, four orders of magnitude improvement over the state-of-the-art methods^9,10,11, while consuming 460× less energy than digital computers^12,13. Accelerators leveraging these molecular crossbars could transform neuromorphic computing, extending it beyond niche applications and augmenting the core of digital electronics from the cloud to the edge^12,13.

人工智能 (AI) 是大型资源密集型数据中心的领域，这些数据中心限制了小型开发人员社区的访问^1,2 。神经形态硬件有望大大提高人工智能的空间和能源效率，但目前只能进行低精度操作，例如神经网络中的推理^3,4,5 。信号处理、神经网络训练和自然语言处理的核心计算任务需要更高的计算分辨率，超出了单个神经形态电路元件的分辨率^6,7,8 。在这里，我们介绍一种基于偶氮芳香族配体 Ru 络合物的模拟分子忆阻器，具有 14 位分辨率。对两个热力学稳定分子电子态之间转变的精确动力学控制有利于 16,520 个不同的模拟电导水平，这些水平可以线性对称地更新或在一个时间步长中单独写入，从而大大简化了现有神经形态平台³的权重更新程序。电路元件是单向的，有利于无选择器 64 × 64 基于交叉的点积引擎，可在单个时间步长中实现矢量矩阵乘法，包括傅里叶变换。我们实现了超过 73 dB 的信噪比，比最先进的方法提高了四个数量级^9,10,11 ，同时消耗的能量比数字计算机少 460 倍^12,13 。利用这些分子交叉开关的加速器可以改变神经形态计算，将其扩展到利基应用之外，并增强从云端到边缘的数字电子核心^12,13 。