Overview

NeuPro™ is a dedicated low power AI processor family for Deep Learning at the edge. Providing a self-contained, specialized AI processors, scaling in performance for a broad range of end markets including IoT, smartphones, surveillance, automotive, robotics, medical and industrial.
NeuPro builds on CEVA’s industry-leading position and experience in deep neural networks for computer vision applications. Dozens of customers are already deploying the CEVA-XM4 and CEVA-XM6 vision platforms along with the CDNN SW Compilerrk in consumer, surveillance and ADAS products. This new family of dedicated AI processors offers a considerable step-up in performance, ranging from 2 Tera Ops Per Second (TOPS) for the entry-level processor and 12.5 TOPS for the most advanced configuration.

Benefits

The NeuPro AI processor family were designed to reduce the high barriers-to-entry into the AI space in terms of both architecture and software. Enabling an optimized and cost-effective standard AI platform that can be utilized for a multitude of AI-based workloads and applications

Self-contained AI Processor - reduce the high barriers-to-entry into the AI space in terms of both architecture and software
Offers a considerable step-up in performance. Ranging from 2 TOPS up to 12.5 TOPS for the most advanced configuration
Optimized for scalable power consumption, performance, and area (PPA) requirements

Main Features

  • NeuPro AI processor consists of the NeuPro Engine and the NeuPro VPU
    • NeuPro Engine - Specialized engines for Matrix Multiplication, Fully Connected, Activation and Pooling layers
    • NeuPro VPU - Fully programmable Vector Processor Unit for customer extensions, customization and CDNN compiler
  • Supports both 8bit and 16bit quantization
    • Optimized real-time decision made in order to achieve the best tradeoff between precision vs. performance
  • Support up to 4K 8x8 MACs
  • Support all layer types and NN topologies
  • Optimized DDR BW
    • Advanced DMA controller
    • On-the-fly Activation and Pooling pipeline processing
  • The NeuPro family comprises four AI processors offering different levels of parallel processing:
    • NP500 is the smallest processor, including 512 MAC units and targeting IoT, wearables and cameras
    • NP1000 includes 1024 MAC units and targets mid-range smartphones, ADAS, industrial applications and AR/VR headsets
    • NP2000 includes 2048 MAC units and targets high-end smartphones, surveillance, robots and drones
    • NP4000 includes 4096 MAC units for high-performance edge processing in enterprise surveillance and autonomous driving

Block Diagram

Microprocessor Report

New IP Comprises a General-Purpose Machine-Learning Processor

NeuPro’s target applications include advanced driver-assistance systems (ADASs), augmented-reality (AR) headsets, drones, smartphones, and surveillance cameras.