NeuReality recently started showing FPGA prototypes of its NR1 server-on-a-chip ASIC (see NeuReality starts showing prototype inference platform) and the agreement focuses on this even though IBM has developed its own dedicated AI chip called Telum (see IBM shows first dedicated AI inference chip).
When implemented in its own dedicated silicon the NR1 will support AI-over-Fabric networking, NeuReality said in a statement. It said the two companies will use the technology to pursue such applications as AI applied to finance, insurance, healthcare, manufacturing, computer vision, natural language processing, recommendation systems, and other AI use cases.
As part of the partnership NeuReality has become the a member of the IBM Research AI Hardware Center and licensee of the Center’s low-precision high performance Digital AI Cores.
IBM will provide cores and become a design partner with NeuReality and work on the NR1 chip specifications, the associated software design kit and what will be implemented in the next revision of the architecture.
Together the two companies will evaluate NeuReality’s products for use in IBM’s Hybrid Cloud, including AI use cases, system flows, virtualization, networking, security and more.
“Having the NR1-P FPGA platform available today allows us to develop IBM’s requirements and test them before the NR1 Server-on-a-Chip’s tapeout,” said Moshe Tanach, CEO and co-founder of NeuReality, in a statement.
The IBM Research AI Hardware Center is a global research hub headquartered in Albany, New York. The center is focused on inventing and enabling next-generation chips and systems that deliver the tremendous processing power and unprecedented speed required for AI to realize its full potential.
Related links and articles: