Weebit Nano Launches Weebit Matrix™ In-Memory Compute Research Program

HOD HASHARON, Israel, Nov. 04, 2026 /CourierPR/ -- Weebit Nano, a leading developer and licensor of advanced memory technologies, has launched the Weebit Matrix™ In-Memory Compute (IMC) Research Program. This initiative aims to accelerate the exploration of next-generation computing architectures by providing industry and academic researchers with a ReRAM-based hardware development kit. The program is designed to facilitate the development and testing of AI algorithms, innovative programming methods, and new AI and neuromorphic architectures.

In modern AI workloads, moving data between the processor and memory can create significant performance and bandwidth bottlenecks, accounting for a substantial part of total energy consumption. IMC, which performs calculations directly within memory, is seen as a promising solution to these challenges. ReRAM, a non-volatile memory technology, is particularly appealing for AI inference applications, especially in edge and IoT devices due to its ultra-low power consumption, scalability, cost-efficiency, analog characteristics, and ease of fabrication in the back end of the line (BEOL).

Coby Hanoch, CEO of Weebit Nano, highlighted the company's long-standing collaboration with partners on in-memory and neuromorphic computing research.

The Weebit Matrix IMC Research Program offers two research-focused packages: the Weebit Matrix IMC Evaluation Package, which includes a direct-access evaluation board, and the Weebit Matrix IMC Quick-Start Package, which includes a complete IMC board, FPGA subsystem, and sample code. Each package includes chip samples, documentation, and an IMC board with a socket. The Weebit Matrix IMC chip features a 32×32 crossbar array architecture designed to support efficient multiply-accumulate (MAC) operations, vector-matrix multiplication, convolution processing, spiking neural networks (SNN), and analog computing.

Weebit will demonstrate the Weebit Matrix™ IMC at its booth at electronica 2026 (Hall C2, stand 637). To learn more or arrange a personal demonstration, visit the company's website. Requests for participation in the research program can be submitted through the provided link, where applications will be evaluated to ensure alignment with the program's research focus.

Kevin D. Johnson, a neuromorphic AI technologist and researcher, expressed his interest in combining ReRAM with other processing technologies.

Weebit Nano's commitment to advancing the technology through collaboration with the research community is evident in its ongoing efforts. The company's ReRAM could consume approximately 80x less power than a GPU-based implementation for a given AI inference workload, while also reducing latency by almost 100x, as demonstrated by a U.S.-based company specializing in radiation-tolerant AI hardware.