CEVA, Inc. (NASDAQ: CEVA), the leading licensor of wireless connectivity and smart sensing technologies and co-creation solutions announced today that NeuPro-M AI Processor architecture for artificial intelligence and machine learning (AI/ML) inference workloads has received industry recognition with ‘Most Promising Product’ at the EE Awards Asia 2022.
‘Creating the future with the electronics industry and changing the world with engineers’ is the cited aim of the EE Awards Asia 2022, with the program garnering considerable interest from leading companies across the electronics industry.
Ran Snir, Vice President and General Manager of the Vision Business Unit at CEVA, commented: “To be recognized for our NeuPro-M AI/ML architecture, among the engineering community is a big honor for CEVA. This is an important validation for NeuPro-M, and evidence that Asia’s engineers understand the value proposition that NeuPro-M brings for low power, high performance edge AI processing, in automotive, industrial, surveillance, smartphones and many more end markets.”
Neural niche
Targeting the broad markets of Edge AI and Edge Compute, NeuPro-M is a self-contained heterogeneous architecture. It is composed of multiple specialized co-processors and configurable hardware accelerators that seamlessly and simultaneously process diverse workloads of Deep Neural Networks, boosting performance by 5-15X compared to its predecessor.
An industry first, NeuPro-M supports both system-on-chip (SoC) as well as Heterogeneous SoC (HSoC) scalability to achieve up to 1,200 TOPS and offers optional robust secure boot and end-to-end data privacy.
Together with CEVA’s multi award-winning neural network AI compiler – CDNN – and its robust software development environment, NeuPro-M provides a fully programmable hardware/software AI development environment for customers to maximize their AI performance. CDNN includes innovative software that can fully utilize the customers’ NeuPro-M customized hardware to optimize power, performance & bandwidth. The CDNN software also includes a memory manager for memory reduction and optimal load balancing algorithms, and wide support of various network formats including ONNX, Caffe, TensorFlow, TensorFlow Lite, Pytorch and more. CDNN is compatible with common open-source frameworks, including Glow, tvm, Halide and TensorFlow and includes model optimization features like ‘layer fusion’ and ‘post training quantization’ all while using precision conservation methods.
NeuPro-M customers can also benefit from SoC design services from CEVA to help integrate and support system design, up to co-creation of the entire SoC. For further information, visit NeuPro-M AI Processor product page.