‘Spatial intelligence shouldn’t require a GPU': This tiny chip could turn skinny Aviators into smart glasses without a bulky battery - and they'll even understand the world in real-time

May 16, 2026 - 22:13
0 1
‘Spatial intelligence shouldn’t require a GPU': This tiny chip could turn skinny Aviators into smart glasses without a bulky battery - and they'll even understand the world in real-time
Mosaic Chip Emulated On FPGA (Image credit: Mosaic SoC)

  • This brand new perception chip supports real-time awareness at huge efficiency
  • Always-on tracking could also improve smartphone camera performance
  • Object recognition, positional tracking and scene understanding also supported

Swiss semiconductor startup Mosaic has successfully raised $3.8 million in funding to build super-efficient so-called perception chips, which would be installed across smart glasses and other wearables to bring more powerful processing into smaller form factors.

Company co-founder and CEO Alfio Di Mauro argues that “spatial intelligence shouldn’t require an application-class processor and a GPU,” hence the inception of Mosaic SoC, which aims to deliver “real-time perception at a fraction of the energy.”

Mosaic says its “next-generation perception chips” will address some of the current bottlenecks wearable makers face today, including battery size, heat output, device thickness and engineering complexity.

Are perception chips the answer to powerful yet slim wearables?

While cameras and sensors continue to improve, manufacturers are under increasing pressure to add more compute and processing power inside devices while maintaining (or even decreasing) form factor.

The perception chip is said to be able to provide real-time environmental awareness, object recognition, positional tracking and scene understanding while consuming only a fraction of the power a conventional smartphone-class compute stack would consume, making it ideal for future AR glasses and headsets.

“The Mosaic SoC chips are designed to be small enough and efficient enough to make smart glasses indistinguishable from regular glasses, while still delivering full spatial awareness,” the company declared in a press release.

Early smart glasses like Ray-Ban Meta and Oakley Meta devices currently lean heavily on smartphone pairing to get the most out of the technology they provide because the physical constraints of size and weight prevent them from being able to install the relevant hardware. On the flip side, Apple’s Vision Pro often draws criticism for its sheer weight (750-800g), which makes it uncomfortable to wear for longer periods.

Sign up to the TechRadar Pro newsletter to get all the top news, opinion, features and guidance your business needs to succeed!

However use cases extend far beyond just glasses – in smartphones, perception chips could power always-on tracking and classification to deliver continuous awareness with minimal impacts on battery life.

Extra power and big goals

Where Mosaic’s perception chip differs from others is that, where others rely on single- or dual-code ARM-based designs, this one uses a proprietary design and a multi-core architecture of eight or more cores, which maximizes performance per watt to back up Mosaic’s bold efficiency claims.

Mosaic also claims to remove original design manufacturer (ODM) complexity, shipping its chips with a full application layer developed and maintained by the startup.

While the company has already achieved “meaningful revenue” through NRE contracts with ODM partners, future plans revolve around far more than being a chip provider. This startup has big visions to become a platform supplier, where applications are developed specifically for its silicon.

“The next billion smart devices will see and understand the world around them,” Antonia Albert, an investor with leading Swiss pre-seed fund Founderful, commented. “Mosaic SoC's product is the chip that makes that possible at scale.”


Google logo on a black background next to text reading 'Click to follow TechRadar'

Follow TechRadar on Google News and add us as a preferred source to get our expert news, reviews, and opinion in your feeds.

With several years’ experience freelancing in tech and automotive circles, Craig’s specific interests lie in technology that is designed to better our lives, including AI and ML, productivity aids, and smart fitness. He is also passionate about cars and the decarbonisation of personal transportation. As an avid bargain-hunter, you can be sure that any deal Craig finds is top value!

What's Your Reaction?

Like Like 0
Dislike Dislike 0
Love Love 0
Funny Funny 0
Wow Wow 0
Sad Sad 0
Angry Angry 0

Comments (0)

User