Mouser Electronics, the NPI leader empowering innovation, announces a new global distribution agreement with Hailo, a leading chipmaker of edge AI processors. Hailo’s unique, patented silicon architecture enables local, private, efficient, high-performance AI processing with low power consumption, small size, and low cost. Their portfolio includes AI accelerators that integrate with edge platforms, enabling real-time deep learning inference at the edge, as well as a range of AI vision processors that integrate Hailo’s leading AI inference capabilities with advanced computer vision engines to empower smart cameras with unprecedented AI capabilities.
Among the Hailo products now available from Mouser is the Hailo-8 edge AI processor, which delivers 26 tera-operations per second (TOPS) and offers industry-leading performance. The Hailo-8 integrates all required memory directly on the processor die, delivering the bandwidth needed for high-performance AI inference while reducing system cost by eliminating the need for external DRAM, simplifying hardware design, and improving reliability. The Hailo‑8 achieves outstanding efficiency in neural network (NN) deployment through distinctive architectural features within its NN Core. These include a distributed memory fabric with specialized pipeline components that enable ultra-low-power memory access during NN processing, highly efficient compute units that can be dynamically allocated based on workload demands, and a dataflow-driven interconnect that adapts to the neural network’s structure, maximizing overall resource utilization. By combining high performance, low power consumption, and advanced video analytics capabilities, Hailo-8 supports a wide range of edge AI applications, running locally on the edge, independent of internet connectivity.
The Hailo-8L entry-level AI accelerator, available to order from Mouser, shares the same innovative architecture and core features as the Hailo-8, while offering a lower performance grade optimized for applications that prioritize cost and power over peak performance. The Hailo-8L delivers 13 TOPS and supports entry-level products requiring limited AI capacity or lower performance. The device boasts exceptional low-latency, high-efficiency processing, capable of handling complex pipelines with multiple real-time streams and concurrent processing of multiple models and AI tasks. Like the Hailo-8, the Hailo-8L integrates all required memory, reducing supply chain risk and BoM costs.
The Hailo-8 M.2 Starter Kits include a Hailo-8 M.2 AI acceleration module and a thermal management kit. The module can be plugged into an existing edge device with an M.2 socket to execute deep neural network inference in real time, consuming low power and serving a broad range of market segments. The Hailo-8 M.2 Starter Kits enable users to quickly and easily integrate the Hailo-8 into existing applications.
The Hailo-10H M.2 AI Acceleration Starter Kits include the Hailo-10H M.2 Key M ET AI acceleration module in an M.2 form factor. The Hailo-10H delivers 40 TOPS of INT4 performance and exceptional power efficiency. It can be added to existing edge devices that feature an M.2 socket. The Hailo-10H includes a direct DDR interface, allowing it to scale for large models such as LLMs, VLMs, and more.
All Hailo products are supported by a comprehensive software suite and a global community of thousands of developers and makers. The Hailo SDK is now compatible with agentic AI, enabling developers to use natural language and agent-based workflows for fast and simple application development.
