NeuReality, an Israeli-based semiconductor company developing high-performance AI inference technology, has signed an agreement with IBM to develop the technology.
The technology aims to deliver cost and power consumption improvements for deep learning use cases of inference, the companies said. This development follows NeuReality’s emergence from stealth earlier in February with an $8 million seed round to accelerate AI workloads at scale.
AI inference is a growing area of focus for enterprises, because it’s the part of AI where neural networks actually are applied in real application and yield results. IBM and NeuReality claim their partnership will allow the deployment of computer vision, recommendation systems, natural language processing, and other AI use cases in critical sectors like finance, insurance, healthcare, manufacturing, and smart cities. They also claim the agreement will accelerate deployments in today’s ever-growing AI use cases, which are already deployed in public and private cloud datacenters.
NeuReality has competition in Cast AI, a technology company offering a platform that “allows developers to deploy, manage, and cost-optimize applications in multiple clouds simultaneously.” Some other competitors include Comet.ml, Upright Project, OctoML, Deci, and DeepCube. However, this partnership with IBM will see NeuReality become the first start-up semiconductor product member of the IBM Research AI Hardware Center and a licensee of the Center’s low-precision high performance Digital AI Cores.
VentureBeat connected via email with Moshe Tanach, CEO and co-founder of NeuReality, to get a broader view on the direction of this partnership.
Delivering a new reality to datacenters and near edge compute solutions
NeuReality’s agreement with IBM includes cooperation around NR1, NeuReality’s first Server-on-a-Chip ASIC implementation of its AI-centric architecture. The NR1 is a high performance, fully linear, scalable, network-attached device that provides services of AI workload processing, NeuReality says. In simpler terms, the NR1 offering targets cloud and enterprise datacenters, alongside carriers, telecom operators, and other near edge compute solutions—enabling them to deploy AI use cases more efficiently. The NR1 is based on NeuReality’s first generation FPGA-based NR1-P prototype platform introduced earlier this year.
In line with NeuReality’s vision to make AI accessible to all, this technology will remove the system bottlenecks of today’s solutions and provide disruptive cost and power consumption benefits for inference systems and services, the company said. The collaboration with IBM will ensure Neurality’s already available FPGA-based NR1-P platform supports software integration and system level validation prior to the availability of the NR1 production platform next year, the companies said.
“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. Being able to develop, test and optimize complex datacenter distributed features, such as Kubernetes, networking, and security before production is the only way to deliver high quality to our customers. I am extremely proud of our engineering team who will deliver a new reality to datacenters and near edge solutions. This new reality will allow many new sectors to deploy AI use cases more efficiently than ever before,” Tanach added.
A marker of NeuReality’s continued momentum
According to Dr. Mukesh Khare, Vice President of Hybrid Cloud research at IBM Research, “In light of IBM’s vision to deliver the most advanced Hybrid Cloud and AI systems and services to our clients, teaming up with NeuReality, which brings a disruptive AI-centric approach to the table, is the type of industry collaboration we are looking for. The partnership with NeuReality is expected to drive a more streamlined and accessible AI infrastructure, which has the potential to enhance people’s lives.”