In summary
- Weebit Nano has launched the Weebit Matrix In-Memory Compute Research Program, giving researchers a ReRAM-based hardware development kit to prototype next-generation computing architectures.
- The platform is offered as two packages, an Evaluation Package with a direct-access evaluation board and a Quick-Start Package with a complete IMC board, FPGA subsystem and sample code.
- Chief executive Coby Hanoch said strong interest from companies and research institutes is already translating into engagements with industry players joining the research program.
Weebit Nano, an ASX-listed developer and licensor of advanced memory technologies for the semiconductor industry, has launched the Weebit Matrix In-Memory Compute (IMC) Research Program, giving industry and academic researchers a ReRAM-based hardware development kit to explore and prototype next-generation computing architectures.
In-memory compute (IMC) performs calculations directly within memory, reducing data transfers and lowering power consumption - a proposition the company said is particularly suited to AI inference workloads on edge and IoT devices, where moving data between processor and memory can create performance bottlenecks and account for a substantial share of total energy use.
The programme gives researchers access to Weebit's ReRAM platform to develop and test AI algorithms, investigate programming methods, explore neuromorphic architectures and assess potential power and performance gains. Participating researchers will share findings and feedback with Weebit to help refine the technology.
Weebit is offering the platform as two packages: a Weebit Matrix IMC Evaluation Package, which includes a direct-access evaluation board; and a Weebit Matrix IMC Quick-Start Package, which includes a complete IMC board, an FPGA subsystem and sample code.
Both packages include chip samples, documentation and an IMC board with a socket.
The underlying chip features a 32×32 crossbar array designed to support multiply-accumulate operations, vector-matrix multiplication, convolution processing, spiking neural networks and analogue computing.
Weebit chief executive Coby Hanoch said the company had been collaborating on in-memory and neuromorphic computing research for years, including a ReRAM-based spiking neural network demonstration with CEA-Leti in 2019.
"We're seeing strong interest from companies and research institutes looking to explore how Weebit ReRAM can support such computing paradigms," Hanoch said.
"This interest is already translating into engagements with industry players looking to participate in the research program."
“While there are still technical challenges to address before bringing a commercial IMC solution to market, this research program will allow us to collaborate with the research community, exchange knowledge and ideas, and help advance the technology.”




