Raspberry Pi OS "Trixie" Gets Hailo-Based AI Kit, AI HAT+ Support — And a New AI Camera Feature

Raspberry Pi has announced an update to its latest Debian 13 "Trixie"-based Raspberry Pi OS, which adds previously-missing support for the company's on-device machine learning and artificial intelligence (ML and AI) accelerator products — both the Raspberry Pi AI Kit and Raspberry Pi AI HAT+.

"The Raspberry Pi AI HAT+ and the Raspberry Pi AI Kit, both based on Hailo AI accelerators, are now fully supported on the recently released Trixie version of Raspberry Pi OS," Raspberry Pi software engineering manager Naush Patuck explains of the new updates. "All the required software packages are available and ready to install from our apt repo."

The new support covers both the Raspberry Pi AI Kit, launched back in June 2024, and the newer Raspberry Pi AI HAT+, which came out in October that year — both before the release of the latest Raspberry Pi OS, which was first made publicly available in July 2025. While the Debian 13 "Trixie"-based operating system brought a range of improvements, it left those experimenting with the AI Kit and AI HAT+ behind — until now.

The updates, available today, now provide full compatibility with both the devices without having to stay behind on a now-unsupported version of Raspberry Pi OS — whether you're using the Hailo-8L variant, which delivers 13 tera-operations per second (TOPS) of INT8-precision compute, or the more powerful Hailo-8 that hits 26 TOPS at the same precision.

"This package release does contain one significant change," Patuck notes. "We have removed the Hailo device driver from our kernel builds and are now using DKMS to build and install the kernel driver as part of the package installation. This decoupling not only enables more flexibility with software releases going forward, but also allows our users to downgrade the device driver without downgrading the kernel itself."

The updates come shortly after Hailo made its new application infrastructure framework available, making it easier to develop custom applications accelerated by the chips — whether on a Raspberry Pi or another Hailo-equipped device. Patuck has also confirmed a new feature for the Raspberry Pi AI Camera, which can run small computer vision models on-sensor: an input tensor injector that provides easier debugging of custom neural networks running on the Sony IMX500 "Intelligent Vision Sensor."

Raspberry Pi OS users with a Raspberry Pi AI Kit or Raspberry Pi AI HAT+ can install the new software with the command sudo apt update && sudo apt install dkms && sudo apt install hailo-all, and will find the new Hailo framework on GitHub; the new input tensor injector for the Raspberry Pi AI Camera is installed automatically with a sudo apt update && sudo apt full-upgrade -y.