The AIE-CP1A-A1 model incorporates the small form factor Blaize® Pathfinder P1600 embedded SoM, ideal for harsh and rugged environments and embedded systems requiring extended temperature ranges. It features dual ARM Cortex A53 processors, 16 GSP cores delivering 16 TOPS AI performance, and the advantages of low memory bandwidth and low latency, enabling it to accelerate the deployment of neural networks (NNs) and deep learning (DL).

Based on Blaize's Graph Streaming Processor (GSP®) architecture, Pathfinder delivers energy-efficient processing power, ideal for edge AI inference workloads. It easily connects to a custom motherboard and can operate independently or be integrated into a larger system. Thanks to its compact, fanless design and multiple I/O ports, the AIE-CP1A-A1 can be adapted to both commercial and industrial AI systems involving vision-related AI inference. The Blaize Pathfinder also supports H.264/H.265 video encoding and decoding, making it compatible with various sensors for fast image recognition and video analytics tasks.

Aetina offers custom motherboards and computer enclosures for sophisticated system integration and chip-down design services to provide other types of embedded system models, such as single-board computers (SBCs) with the Blaize Pathfinder, in addition to standard desktop PCs. Furthermore, Aetina helps developers convert their AI inference models into a format that can run and function smoothly on edge devices built with ASIC chips.

“We are expanding our edge computing product line to offer more GPU- and ASIC-based solutions that help developers build their AI systems across various industries and verticals,” says Jackal Chen, senior product manager at Aetina. “The AIE-CP1A-A1 model, powered by the Blaize Pathfinder P1600 SoM module, is our first ASIC-based system. In the future, by integrating different types of chips and modules, we will be able to offer heterogeneous computing devices suitable for various AI models.”

“Keeping AI processing and inference workloads at the edge, rather than sending data to the cloud, is critical for cost-effective, near-zero-latency AI applications,” says Dinakar Munagala, CEO and co-founder of Blaize. “The Pathfinder SoM and our GSP architecture enable this key business objective for the demands of computer vision applications, including object detection, autonomous optical inspection, robotics, and more.”

In addition, Aetina is now developing more ASIC-based AI accelerators with M.2 and EDSFF specifications for developers, speeding up the inference processes of their AI models at the edge.