Anker Unveils Thus AI Chip: World's First Neural-Net Compute-in-Memory Audio Processor for Local AI in Devices

Anker has announced the development of its own custom silicon, the Thus processor, designed to integrate local AI capabilities into audio devices, mobile accessories, and IoT products. The chip is positioned as the world's first neural-net compute-in-memory AI audio processor, offering a smaller form factor and reduced power consumption compared to conventional chips. This efficiency enables complex computations to run on smaller devices without compromising performance.

In a statement, Anker CEO Steven Yang highlighted the innovation behind Thus, noting,

"Every AI chip built until now stores the model on one side and does the computation on the other. To think, the device has to carry all those parameters across, many times per second, every single inference."

The Thus processor leverages compute-in-memory technology, which integrates memory and processing units to minimize data transfer delays and energy usage. This approach is particularly beneficial for battery-powered and compact devices, where power efficiency and space constraints are critical.

Anker has not disclosed specific technical specifications such as clock speeds, power consumption figures, or detailed performance benchmarks for the Thus chip. However, the company emphasizes its potential to enhance AI-driven features in a wide range of consumer electronics, including smart speakers, wireless earbuds, and connected home devices.

The announcement underscores Anker's strategic push toward proprietary hardware innovation, complementing its existing ecosystem of chargers, batteries, and audio products. The Thus processor is expected to debut in future Anker devices, though no timeline for commercial release has been provided.