Introducing the GW16168 AI Accelerator
In a significant leap for embedded designs,
Ezurio has unveiled its latest innovation—the
GW16168. This advanced AI accelerator, designed for the
M.2 M-Key 2280 form factor, marks a major shift in how artificial intelligence is integrated into devices. With the growing trend of moving AI workloads from the cloud to local devices, the GW16168 is poised to revolutionize various industrial applications.
Key Features
The GW16168 is powered by the state-of-the-art
NXP Ara240 Discrete Neural Processing Unit (DNPU), capable of delivering up to
40 equivalent tera operations per second (eTOPS) of AI acceleration. Notably, this accelerator enables device-side AI inference without requiring any cloud connectivity, making it ideal for applications in challenging environments. Whether it's drone technology, robotics, or remote industrial systems, the GW16168 ensures uninterrupted performance and real-time processing capabilities.
Tailored for Demanding Applications
Made
exclusively in the USA, the GW16168 brings essential transparency and quality assurance. It integrates seamlessly with
VeniceFLEX and Catalina single board computers while maintaining compatibility with other M.2 M-Key 2280 platforms. Designed to withstand extreme conditions, it operates within an
industrial temperature range of -40°C to +85°C, making it exceptionally reliable for critical applications.
Accelerated Development with GW11062
To facilitate developers, Ezurio offers the
GW11062 evaluation kit, which pairs the GW16168 accelerator with a Catalina SBC. This kit reduces setup hassles by including a preloaded
Ubuntu image, allowing rapid testing and adoption. With configurations available for 8GB and 16GB, developers can easily evaluate the GW16168's performance.
A Shift towards Edge AI
According to Eric Singer,
Ezurio's Product Manager, the landscape of edge AI is evolving rapidly from pilot phases to full-scale production. The GW16168 provides a dependable hardware solution that customers can trust. By merging
NXP's Ara240 DNPU technology with a robust industrial-grade design, users can elevate their AI capabilities without overhauling their existing platforms. This flexibility not only enhances deployment efficiency but also supports long-term operational sustainability.
In addition to its exceptional on-device AI processing, the GW16168 is equipped with
features such as:
- - Secure Boot and Root of Trust
- - Support for TensorFlow, PyTorch, and ONNX models via NXP Ara SDK
- - PCIe Gen4 x4 interface for rapid data transfer
- - Long lifecycle commitment ensuring continued support and availability
Conclusion
The launch of the GW16168 AI accelerator is more than just a product release; it signifies an essential evolution in embedded AI technology. By enabling robust data processing directly on devices, Ezurio is championing a new era for industries that rely heavily on AI—allowing for smarter, faster, and more secure operations.
For those seeking to integrate advanced AI into their operations without cloud dependencies, the GW16168 presents an unparalleled opportunity.
For further details, visit
Ezurio's official page.