Description

Waveshare Hailo-8 M.2 AI Accelerator Module Suitable for Raspberry Pi 5 is a compact edge AI acceleration kit designed to bring high-performance neural network inference to the Raspberry Pi 5. The kit combines a Hailo-8 AI M.2 module with a PCIe to M.2 adapter, making it easier to add dedicated AI compute to Raspberry Pi based machine vision, robotics, home automation, process control, and edge intelligence projects.

At the core of this kit is the Hailo-8 AI processor, delivering up to 26 TOPS of AI performance with high power efficiency and low latency. The Hailo-8 M.2 accelerator supports PCIe Gen3 x4, works with mainstream frameworks such as TensorFlow, TensorFlow Lite, ONNX, Keras, and PyTorch, and helps edge devices run multi-stream, multi-model AI workloads more efficiently.

Features

• Powered by the Hailo-8 AI processor with up to 26 TOPS AI performance.
• Typical power consumption of 2.5W for efficient edge AI deployment.
• M.2 Key M form factor with PCIe Gen3 x4 interface for high-speed data transfer.
• Supports TensorFlow, TensorFlow Lite, ONNX, Keras, and PyTorch frameworks.
• Supports Linux and Windows environments.
• Designed for real-time, low-latency, and high-efficiency AI inference at the edge.
• Scalable for multi-stream and multi-model workloads, making it suitable for more demanding AI applications.
• PCIe to M.2 adapter is Raspberry Pi HAT+ compliant for convenient Raspberry Pi 5 integration.
• The adapter includes power monitoring chip and EEPROM for more stable operation and real-time device power monitoring.
• Reserved airflow vent supports adding a cooling fan for better heat dissipation and improved AI performance stability.

Specifications

Applications

• Raspberry Pi 5 AI acceleration for local neural network inference.
• Machine vision and smart camera projects with real-time edge inference.
• Robotics systems that need efficient onboard AI processing.
• Home automation and intelligent control applications.
• Process control and industrial edge computing workloads.
• Multi-stream AI systems that require scalable inference performance.

Product List

• Hailo-8 AI M.2 module 1
• PCIe to M.2 adapter 1

Other Content

Edge AI performance upgrade: the Hailo-8 M.2 accelerator adds dedicated neural network inference capability to existing edge devices with an M.2 socket, helping them run AI models with lower latency and higher efficiency than CPU-only processing.

Raspberry Pi 5 integration: with the included PCIe to M.2 adapter, this kit is especially practical for Raspberry Pi 5 AI projects that need compact installation, stable power monitoring, and expandable cooling options.

Related reading on OpenELAB: What Is the Raspberry Pi AI Kit, How to Set Up Raspberry Pi 5 Hailo AI Kit and Deploy YOLO Object Detection.

Resources

• Waveshare Wiki
• Hailo-8 AI Accelerator Overview
• Hailo Software Downloads
• Raspberry Pi PCIe Documentation
• Raspberry Pi 5 Platform Overview

FAQ

1. Is this only an AI module or a complete kit?
This SKU is the Kit version, including the Hailo-8 AI M.2 module and the PCIe to M.2 adapter.

2. What is the AI performance of the Hailo-8 module?
The Hailo-8 AI processor delivers up to 26 TOPS of AI performance.

3. Which frameworks does it support?
It supports TensorFlow, TensorFlow Lite, ONNX, Keras, and PyTorch.

4. What kinds of projects is it suitable for?
It is suitable for Raspberry Pi 5 AI projects, machine vision, robotics, home automation, and industrial edge AI.