AI Accelerator IP – ENLIGHT™

OPENEDGES™ Artificial Intelligence Compute Engine
 ENLIGHT™ is a   deep learning accelerator IP technology delivers unrivaled compute density and energy efficiency.  ENLIGHT™ NPU IP accelerates inferencing computation with excellent efficiency in terms of power/performance/area. It targets  +10x compute density (performance/area)  and +10x power efficiency (performance / power) vs comparable GPU.ENLIGHT™ optimize precision for common case and then deal with outliers somehow. The key observations are that most of the weights (& activations) can be nicely quantized in 4-bits and higher-precision outliers account for less than 3% of overall weights count.
ENLIGHT™ implementation an array of 4-bit precision MAC unit and it enables 4x compute density & low-power consumption.

Also our ORBIT™ high performance memory subsystem IP can perfectly meet high DRAM bandwidth requirement.

ENLIGHT

 

ENLIGHT™ NPU
: Deliver unrivaled compute density and energy efficiency.

ENLIGHT™ Neural Network Compiler
: Optimize your neural network for perfect fit to ENLIGHT™ NPU.

ORBIT™ Memory Subsystem
: Pump exceptional memory performance into the ENLIGHT™ NPU and the rest of the system.