Lanner Electronics, a global leader in the design and manufacturing of intelligent edge computing appliances, announced its first Hailo-8™AI-powered PCIe accelerator card, the Falcon H8. Lanner collaborated with leading AI (Artificial Intelligence) chipmaker Hailo to design the Falcon H8, enabling scalable and powerful intelligent video analytics applications for multiple industries operating at the edge, including intelligent transport systems (ITS), smart cities, smart retail, and Industry 4.0. The Falcon H8 is one of the most cost-efficient PCIe AI accelerator cards on the market, with a low power consumption and record high of up to 156 tera operations per second (TOPS) to allow high-end deep learning applications on edge servers.
Lanner‘s Falcon H8 modular, PCIe FHHL form factor provides a compact and easily deployable solution for engineers looking to offload CPU loading for low-latency deep learning inference. With high-density AI processors, the Falcon H8 accommodates 4, 5, or 6 Hailo-8™ AI processors, offering a modular, cost-effective Edge AI solution with high processing capabilities and power efficiency. Through a standard PCIe interface, the Falcon H8 AI Accelerator Card enables legacy devices such as NVRs, Edge AI boxes, Industrial PCs and robots to run video-intensive, mission-critical Edge AI applications such as video analytics, traffic management, access control, and beyond.
The Falcon H8 delivers unprecedented inference processing of over 15,000 Frames Per Second (FPS) for MobileNet-v2 and 8,000 FPS for ResNet-50. Its performance is up to 4x more cost effective (TOPS/$) and 2x more power efficient (TOPS/W) compared to leading GPU-based solutions.
“Optimized for AI functionality, performance, and ease of deployment, Lanner is pleased to partner with Hailo to design a next-gen AI accelerator card that brings top-performing AI computing to the edge of industrial IoT,” said Jeans Tseng, CTO of Lanner Electronics. “Our expertise in creating high-density hardware platforms, combined with Hailo’s state-of-the-art neural chip and software framework, provides service providers and system integrators a best-in-class AI accelerator that enables running deep learning applications most efficiently with the lowest total cost of ownership.”