Design-Reuse

OPENEDGES To Exhibit its 4-/8-bit mixed-quantization NPU IP at Embedded Vision Summit 2022

The ENLIGHT, a high-performance neural network processor IP, is the industry’s first adoption of 4-/8-bit mixed-quantization IP and features a highly optimized network model compiler that reduces DRAM ...
Design-Reuse

OPENEDGES Announces the Industry First 4-/8-bit Mixed-Precision Neural Network Processing Unit IP

OPENEDGES Technology, Inc., the world's leading memory system and AI platform IP provider, today announced the first commercial release of mixed-precision (4-/8-bit) computation NPU IP, ENLIGHT™.
VentureBeat

Nvidia researchers unlock 4-bit LLM training that matches 8-bit performance

Researchers at Nvidia have developed a novel approach to train large language models (LLMs) in 4-bit quantized format while maintaining their stability and accuracy at the level of high-precision ...
Semiconductor Engineering

Neural Network Model Quantization On Mobile

The general definition of quantization states that it is the process of mapping continuous infinite values to a smaller set of discrete finite values. In this blog, we will talk about quantization in ...