MemryX, Edge Impulse unite to accelerate edge AI development and deployment

MemryX has designed edge AI chips to serve a variety of industries, including transportation, IoT, robotics, industrial automation and edge computing. These chips can handle heavy workload processing with high performance and low latency. The company uses its in-house at-memory computing and dataflow architecture. Clients can use the Edge Impulse machine learning platform to train AI models on MemryX edge processors through this partnership.

“This new partnership opens up exciting new possibilities for all AI developers,” said Zach Shelby, the co-founder and CEO of Edge Impulse.

MemryX’s AI accelerators work with numerous Artificial Intelligence software frameworks, including PyTorch, TensorFlow, ONNX and Keras. Moreover, they are compatible with market-leading processor architectures, including x86, Arm, and RISC-V. Clients may rely on Linux, Windows or Android systems to generate production-ready models. According to the company, by leveraging the power of MemryX AI accelerators, users can scale AI performance to any desired level.

“Combining Edge Impulse’s abilities in data collection and low-profile algorithm generation and deployment with MemryX’s advanced hardware provides fascinating potential with a wide range of use cases,” added Shelby.

Launched in 2019, MemryX is a venture co-founded by Dr. Wei Lu of the University of Michigan’s Electrical Engineering and Computer Science faculty with over ten years of experience in memory computing architecture.

Earlier this year, Edge Impulse collaborated with BrainChip to support BrainChip’s neural processor AI IP. Users can leverage Edge Impulse’s machine learning platform and BrainChip’s Akida processor chip for edge application deployments.

Article Topics

AI/ML  |  chip  |  development tools  |  edge AI  |  Edge Impulse  |  MemryX  |  model development  |  PyTorch  |  Tensorflow