NPU IP for Embedded ML

• Fully programmable to efficiently execute Neural Networks, feature extraction, signal processing, audio and control code
• Scalable performance to meet a wide range of use cases, with MAC configurations up to 64 int8 (native 128 of 4×8) MACs per cycle
• Two NPU configurations to address a wide variety of use cases are available:
  • Ceva-NPN32 with 32 4×8, 32 8×8, 16 16×8, 8 16×16, 4 32×32 MAC operations per cycle
  • Ceva-NPN64 with 128 4×8, 64 8×8, 32 16×8, 16 16×16, 4 32×32 MAC operations per cycle and 2x performance acceleration using 50% weight sparsity (Sparsity Acceleration)
• Future proof architecture that supports the most advanced ML data types and operators, including 4-bit to 32-bit integer support and native transformer computation
• Ultimate ML performance for all use cases, with Sparsity Acceleration, acceleration of non-linear activation types, and fast quantization – up to 5 times acceleration of internal re-quantizing tasks
• Powerful microcontroller and DSP capabilities with a Coremark/MHz score of 6.0
• Ultra-low memory requirements achieved with Ceva-NetSqueeze™, yielding up to 80% memory footprint reduction through direct processing of compressed model weights without the need for an intermediate decompression stage. NetSqueeze solves a key bottleneck inhibiting the broad adoption of AIoT processors today
• Ultra-low energy achieved through innovative energy optimization​​s, including dynamic voltage and frequency scaling support tunable for the use-case​, and dramatic energy and bandwidth reduction by distilling computations using weight-sparsity acceleration
• Complete, simple to use Ceva-NeuPro-Studio AI SDK, optimized to work seamlessly with leading, open-source AI inference frameworks, such as LiteRT for Microcontrollers and µTVM
• Model Zoo of pre-trained and optimized machine learning models covering Embedded ML audio, voice, vision and sensing use cases
• Comprehensive portfolio of optimized runtime libraries and off-the-shelf application-specific software

Edge AI Sensing eBook or Webinar On-Demand: What it Really Takes to Build a Future-Proof AI Architecture

Ceva-NeuPro-Nano is a fully stand-alone neural processing unit (NPU), not an AI accelerator, and does not require a host CPU or DSP. The IP core includes all processing elements of a self-contained NPU, including code execution and memory management.

Key architectural capabilities:

• Fully programmable architecture for executing neural networks, feature extraction, control code, and DSP code
• Supports advanced ML data types and operators including native transformer computation, sparsity acceleration, and fast quantization
• Efficiently executes a wide range of neural networks with high performance

Two cores are available:

• Ceva-NPN32
• Ceva-NPN64 (includes hardware sparsity acceleration for 2× effective performance)

Additional advantages:

• Hardware-based weight decompression reduces memory footprint by up to 80%
• Full support through Ceva-NeuPro Studio, enabling importing, compiling, and debugging models from frameworks such as LiteRT and µTVM

The Ceva-NeuPro-Nano NPU family is specially designed to bring the power of AI to the Internet of Things (IoT), through efficient deployment of Embedded ML models on low-power, resource-constrained devices. Ceva-NeuPro-Nano NPUs’ optimized, self-sufficient architecture enables them to deliver superior power efficiency, with a smaller silicon footprint, and optimal performance for Embedded ML workloads, compared to the existing processor solutions, which utilize a combination of CPU or DSP with a separate AI accelerator.

The Ceva-NeuPro Studio, which leverages open source AI framework and provides an easy-to-use software development environment, extensive pre-optimized models in the Ceva Model Zoo, and a wide range of runtime libraries, speeds product development for chip designers, OEM’s and software developers.

• Consumer IoT
• Automotive
• Industrial Automation

The EDGE AI FOUNDATION (formerly tinyML Foundation) is a global non-profit community of innovation, collaboration, advocacy and education for efficient, affordable and scalable Edge AI technologies.