Hailo joins Lanner Electronics to Launch the Next Generation of End-Point AI Processing Solutions
Tel Aviv-based startup Hailo, a leading maker of AI chips, today announced a partnership with Taiwan’s Lanner Electronics, a leading global developer and manufacturer of smart endpoint computing devices. As part of the collaboration, the Hailo-8TM AI acceleration module has now been integrated into Lanner Electronics’ computerized end devices to meet the growing demand for small, high-performance devices that allow new AI applications to run at endpoints.
Lanner Electronics provides smart computing devices for endpoints to leading global companies including: Intel, Infinion, Broadcom, Verizon, Microsoft, Mellanox, Red, NEC and other companies.
Smart cities, retail chains and manufacturing plants operate hundreds of cameras that produce video streams that need to be processed locally efficiently and quickly and without delays. Harnessing AI to meet the challenge requires high performance and solutions that can be deployed quickly – as offered by Halo and Lanner Electronics’ integrated solution.
Lanner’s LEC-2290 end device combines an x86 processor with the Hailo-8TM processor AI module, which together create a solution for optimizing demanding loads at the endpoints, shortening the time it takes for the market of AI applications. LEC-7242, the low-power Lanner gateway, now integrates the Hailo-8TM which is not equipped with a fan, thus creating an efficient device for real-time applications for AI processing at endpoints. Each of the two solutions is capable of processing multiple video streams in real time and securely sends the insights and metadata resulting from multiple sources through a communication network installed in the device itself.
Lanner’s tiered access to linked devices provides a clear path for engineers seeking to optimize loads on neural networks. Lanner’s AIs for endpoints that incorporate Halo’s AI module enable critical applications such as video analytics, car traffic management, access control and other applications in real time and maximum efficiency.
The Hailo-8TM AI acceleration module delivers unprecedented performance to end devices. With a processing speed of 26 terabytes per second (TOPS) and innovative architecture, it allows end devices to run deep learning applications that until now could only be run in the cloud. Its advanced structure provides higher performance, lower energy consumption and minimal lag. In doing so, it allows for a deeper protection of the privacy and enhanced reliability of the smart devices operating at the endpoints.