The Raspberry Pi’s new add-on board has 8GB of RAM to run general AI models

Raspberry Pi is launching a new add-on board capable of running generative AI models locally on the Raspberry Pi 5. Announced on Thursdaythe $130 AI Hat + 2 It is an upgraded – and more expensive – version of the unit Launched last yearnow offering 8GB of RAM and a Hailo 10H chip with 40 TOPS of AI performance.

Once connected, the Raspberry Pi 5 will use the AI ​​HAT+ 2 to handle AI workloads while leaving the mainboard’s Arm CPU available to complete other tasks. Unlike the previous AI HAT+, which focused on image-based AI processing, the AI ​​HAT+ 2 comes with built-in RAM and can run small AI models such as the Llama 3.2 and DeepSeek-R1-Distill, along with a series of Qwen models. You can train and fine-tune AI models using the device as well.

A The demo was shared by Raspberry Pi It shows how you can use the add-on panel to run an AI model that is able to generate a textual description of the camera stream, as well as answer the question of whether there are any people in the scene. Another shows how you Text can be translated From French to English with Qwen2.

However, technology YouTubers He found Jeff Girling The standalone Raspberry Pi 5 with 8GB of RAM generally outperforms the AI ​​HAT+ 2 across supported models. Geerling links the lower performance to power draw, as the Pi 5 can operate at up to 10W, while the AI ​​HAT+ 2 is limited to 3W.

Although the AI ​​HAT+ 2 may be ideal for specific use cases, Gerling points out that the additional 8GB of RAM for the add-on board “is not quite enough to give this HAT an edge over just paying for the larger 16GB Pi with more RAM, which will be more flexible and run models faster.” The AI ​​HAT+ released last year can also handle AI image processing at a starting price of $70 cheaper.

More “larger” AI models “are being prepared for updates, and should be available for installation shortly after launch,” Raspberry Pi says.