QNX, a division of BlackBerry, announced support for the Hailo-8 AI Accelerator on QNX Software Development Platform (SDP) 8.0, expanding the hardware options available to customers building AI-powered Edge systems where performance, reliability, and predictability are critical.
The combination of Hailo’s Edge AI acceleration and QNX’s trusted real-time software enables developers to build AI-powered systems for safety- and mission-critical applications where performance and predictability must go hand in hand. Hailo’s technology is deployed across automotive, industrial automation, robotics, and smart infrastructure applications worldwide.
Enabling the next wave of Physical AI
As AI capabilities increasingly move from the Cloud to the Edge, organisations are under growing pressure to deliver intelligent systems that can operate safely and reliably in dynamic real-world environments. Whether building next-generation robots, medical devices, industrial systems, or software-defined vehicles, developers need a joint hardware-software foundation capable of delivering both advanced AI performance and deterministic real-time behaviour.
By combining Hailo’s Edge AI acceleration with QNX’s trusted real-time software foundation, developers can accelerate the deployment of Physical AI systems without compromising the reliability, responsiveness, and predictability required for real-world operations.
Benchmarking predictable AI performance at the Edge
To demonstrate the benefits of running Edge AI workloads on Hailo hardware with QNX, the companies conducted a benchmarking exercise using a Raspberry Pi 5 driving the Hailo-8 AI Accelerator, running on both QNX SDP 8.0 and a real-time Linux environment. The testing showed that AI workloads running on the QNX OS achieved up to 14x greater performance consistency, 2.6x tighter latency distribution, 4.1% higher throughput, and 3.9% lower average latency. Together, the results demonstrate how QNX and Hailo can help developers build AI-powered systems that are not only fast, but also more predictable and reliable in real-world operations.
“Increasingly, advanced embedded devices are looking to AI to unlock new functionality,” said Grant Courville, SVP, Products and Strategy, QNX. “Together, QNX and Hailo are helping customers turn that potential into reality by bringing advanced AI capabilities to the Edge on a trusted software foundation.”
“As demand for intelligent Edge systems continues to grow, developers are looking for technologies that simplify the path from innovation to deployment,” said Max Glover, Chief Revenue Officer at Hailo. “Together with QNX, we’re enabling customers to bring advanced AI capabilities to a broader range of embedded applications and accelerate the next generation of intelligent products.”
Support for the Hailo-8 AI Accelerator expands the range of AI-enabled hardware available on QNX, giving developers greater flexibility to build intelligent systems that combine advanced Edge AI capabilities with the real-time performance, reliability, and determinism required for deployment in the physical world.
QNX provides high-performance foundational software that helps simplify the most complex challenges in industries such as robotics, automotive, medical devices, industrial controls, commercial vehicles, rail, and aerospace and defence. QNX empowers organisations to unlock new possibilities in areas like high-performance computing at the Edge, standards-based virtualisation technologies, and Cloud enablement.
Additional technical details are available in the whitepaper, QNX for Deterministic Physical AI Workloads.
