Raspberry Pi has launched additional AI-focused add-ons to enhance its capabilities for machine learning and AI projects.

The company has been launching new products at a very fast rate lately. This week, the company will add several new products to the lineup aimed at extending the functionalities of the Raspberry Pi 5.
Raspberry Pi has launched additional AI-focused add-ons to enhance its capabilities for machine learning and AI projects.

The company has been launching new products at a very fast rate lately. This week, the company will add several new products to the lineup aimed at extending the functionalities of the Raspberry Pi 5.

Raspberry Pi is mainly recognized for miniature, low-cost single-board computers that are used quite extensively by hobbyists as well as in schools. Industrial and electronics manufacturing companies also employ the usage of Raspberry Pi. The company's flagship product, the Raspberry Pi 5 features an exposed PCIe 3.0 interface, but it only comes with a 16-pin connector.

The company had been selling M.2 HAT+ extension cards that translate the 16-pin connector to a more conventional M.2 connector. HAT stands for "Hardware Attached on Top", cute acronym the company has been using to refer to extension cards you attach to a regular Raspberry Pi.

Users of Raspberry Pi have taken advantage of that M.2 slot to add NVMe SSDs (much more on that below) as well as other add-ons. For instance, Raspberry Pi just last June announced its AI Kit that is pretty much an M.2 extension card with Hailo's neural network inference accelerator.

This is the launch of the company's brand-new HAT+ add-on board with its inference accelerator built in. Available variants of the Raspberry Pi AI HAT+ are at 13 and 26 tera-operations per second, priced at $70 and $110, respectively. The 13 TOPS version makes use of the same module used by the currently available AI Kit.

You will not train a GPT on a Raspberry Pi, but these AI add-ons are very economically worthwhile for doing inference at the edge.

Although the M.2 HAT+ has an M.2 interface, the AI HAT+ is actually a one-chip package for Hailo's inference modules.

Here's what the AI HAT+ (left) looks like versus the M.2 HAT+ when used with the AI Kit (right):

If you have some knowledge regarding PC components, you probably know how most modern NVMe storage drives utilize M.2 connectors. Nonetheless, there are several types of SSDs, and they differ from one another with regard to form factor and the speed in which performance is carried out.

For Raspberry Pi 5, you can use any NVMe SSD that works with PCIe 3.0 and is in the 2230 and 2242 form factors. Off-the-shelf SSDs are going to be just fine with a Raspberry Pi 5 with an M.2 HAT+ extension.

The company also introduces its branded M.2 NVMe SSDs in two configurations-the 256GB sells for $30, and 512GB is available for $45. To put this into perspective, a 256GB SSD with similar specs available in the market costs anywhere between $20 and $30 on Amazon.

The company sells the SSD Kit bundles, both with an M.2 HAT+ along with the same model of SSD in one box. The 256GB version is sold for $40, and the 512GB one for $55.
 
These SSD Kits won't be a revolution in the Raspberry Pi ecosystem, but they are a nice way to ensure that you are buying an SSD compatible with the Raspberry Pi 5 and its M.2 HAT+ attachment.

Blog
|
2024-10-24 19:28:47