The Raspberry Pi AI HAT+ is an advanced add-on board for the Raspberry Pi 5, featuring a built-in Hailo AI accelerator that transforms it into a powerful, energy-efficient AI platform. Available in 13 or 26 TOPS variants, the HAT+ supports a wide range of AI applications, from object detection and facial recognition to image segmentation and pose estimation, making it ideal for process control, robotics, home automation, and more. Fully integrated with Raspberry Pi’s camera software stack, it leverages the PCIe Gen 3 interface for fast data transfer and seamless compatibility with frameworks like TensorFlow and PyTorch.

Its neural processing capabilities optimize real-time tasks, running multiple AI models concurrently with low latency and minimal power consumption. Designed for simplicity, the HAT+ comes with a stacking header and spacers for easy installation, even with an active cooler. It is compatible with all camera modules and the latest Raspberry Pi OS (Bookworm). With scalable architecture, robust thermal performance, and support for existing tools like libcamera and rpicam-apps, the AI HAT+ empowers developers, hobbyists, and educators to innovate in AI-driven projects, such as intelligent surveillance, smart home devices, and robotics, taking Raspberry Pi-based AI applications to new heights.

Specification:

·       Hailo-8 or Hailo-8L accelerator offering 26 TOPS or 13 TOPS inferencing performance respectively

·       FPC/FFC length is 30mm and preassembled on the HAT too

·       Supports single-lane PCIe 2.0 interface (500MB/s peak transfer rate)

·       Conforms to the Raspberry Pi HAT+ Specification

·       Supports installation of the official active cooler.

·       Minimalist design, easy to install

·       The HAT+ comes with LED indicator for power

·       HAILO AI Accelerator:

·       2 variants:

·       HAILO 8L: Offer 13 Tera (Trillion) Operations Per Second (TOPS) AI Processing

·       HAILO 8: Offer 26 Tera (Trillion) Operations Per Second (TOPS) AI Processing

·       Interface: PCIe Gen 3.0, 2-lanes

·       Supported host architectures: X86 or ARM

·       Supported OS: Linux, Windows. In this case, Raspberry Pi OS is fully support

·       Supported AI frameworks: TensorFlow, TensorFlow Lite, Keras, PyTorch & ONNX

·       Best-in-class power efficiency

·       Operating Temperature: 0℃ to 50℃ (ambient)

·       Work out of the box with Raspberry Pi 5 and Raspberry Pi OS

·       Fully integrated with Raspberry Pi Camera Software Stack, ready for computer vision development and deployment.

·       Dimension: 65.0mm x 56.5mm x 5.6mm (max)