Artificial Intelligence (AI) is all the rage these days, and the cost of entry for experimenting with AI Large Language Models (LLMs) has dropped significantly over time. In fact, you can even install multiple LLMs on the small and inexpensive Raspberry Pi 5. Raspberry Pi offers its own AI kit, an M.2 HAT with a Hailo AI acceleration module, but that's not the only game in town. There are a few other ways to integrate powerful AI into your Raspberry Pi 5.
Raspberry Pi AI kit
The Raspberry Pi AI Kit bundles the Raspberry Pi M.2 HAT+ with a Hailo AI accelerator module for use with Raspberry Pi 5. It provides an accessible, cost-effective and energy-efficient way to integrate powerful AI.
Your AI experiment needs could be met on the Raspberry Pi 5
The Raspberry Pi 5 AI Kit is an M.2 HAT+ with a Hailo-8L Neural Processing Unit (NPU). The NPU does the heavy lifting for your AI tasks and the module itself connects to the PCIe 2.0 interface, significantly reducing any I/O bottlenecks that might occur when using a slower USB port.
AI performance is measured in tera operations per second (TOPS). The more TOPS the NPU can offer, the better AI performance you will see. To provide a few points of reference before we delve into the capabilities of the Raspberry Pi 5, Microsoft requires select computers to have an NPU with 40 TOPS or greater in order to carry the Copilot+ branding. The iPhone 16's NPU delivers 35 TOPS, another point of comparison.
The Hailo-8L offered in the Raspberry Pi 5 AI Kit is not quite as robust. It is only capable of reaching 13 TOPS, which is still powerful enough to experiment with multiple AI models. The other popular AI chip for SBCs like the RPi, the Coral Edge, can only achieve 4 TOPS.
Of course, you may want flexibility and choice in how you run your AI experiments. You may also want to use this valuable HAT space for more than just your NPU, such as adding NVMe storage. This is not possible with the Raspberry Pi 5 AI Kit, but other options offer more possibilities. Here are some of our favorites.
4 The Pineboards Ai Bundle (Hailo-8L)
Everything the Raspberry Pi 5 AI Kit offers and more
While the official Raspberry Pi 5 AI Kit only offers the Hailo 8L chip, removing the ability to also add NVMe storage on the PCIe bus, Pineboards offers a way to avoid this compromise. The Pineboards Ai Bundle features the same Hailo 8L chip from the official kit and more.
In addition to the NPU, this board also has an M.2 M-Key connector for NVMe storage. It is compatible with all popular NVMe SSD size options, including M.2 2230, 2242 and 2280 M-Key storage modules. This HAT basically combines the abilities of the Pineboards HatDrive Nano with the same AI chip used in the official Raspberry Pi AI kit.
This option also requires no compromise on AI performance. The included Hailo-8L AI accelerator still offers 13 TOPS of performance, even though the board doubles the performance of the HAT.
The Pineboards Ai Bundle also works with the same software stack used by the Raspberry Pi 5 AI Kit. So if you have already started developing your projects in this software, you can easily switch to Pineboards HAT without losing development time.
Pineboards Ai Bundle (Hailo-8L)
If you want to use your Raspberry Pi 5 with a Hailo 8L accelerator but still have NVMe boot storage, this is the answer. The Pineboards Ai Bundle includes the same Hailo 8L AI chip as in the official Raspberry Pi 5 AI kit, as well as a second M.2 slot for an NVMe SSD module.
3 The Pineboards hat! AI Coral Edge TPU package
A less powerful NPU, sure, but maybe that's all you need… and it still includes NVMe storage
Next, you may have already developed your AI project with the Coral Edge TPU and want to continue this with the Raspberry Pi 5. Pineboards has several Coral Edge HATs, including a HatDrive variant. The Coral Edge offers this TPU together with a second M.2 slot for an SSD module.
This HAT supports an NVMe drive, but only in sizes 2230 and 2242. While this means less storage capacity and potentially higher prices than a 2280 SSD, it is still a good option for many projects. Of course, these are also the only two sizes supported by the Raspberry Pi's official M.2 NVMe HAT, so you're still combining two HATs in one.
As already mentioned, the Coral Edge is significantly less powerful than the Hailo-8L. You only get 4 TOPS of AI acceleration, but that's still more than enough for many machine vision and robotics projects.
Hats AI! Coral TPU Bundle for Raspberry Pi 5
The Pineboards hat! Ai! The Coral Edge TPU Bundle allows you to integrate the Google Coral Edge TPU into your Raspberry Pi 5 projects. It mounts under your Raspberry Pi 5 so it doesn't block access to the GPIO headers.
2 The Pineboards Has AI! Dual Edge Coral TPU package
Like the regular Coral Edge, but twice as big
Pineboards also offers a Coral Dual Edge HAT for the Raspberry Pi 5. This effectively allows you to double your Coral Edge capabilities without the need for a second device. This is important because you can only use one HAT at a time on the Raspberry Pi 5.
The Coral Dual Edge module, like the regular Coral Edge, is an M.2 device that connects to the HAT. The HAT is then connected to the PCIe port on the Raspberry Pi 5.
This board provides your Raspberry Pi 5 with 8 TOPS of AI processing, making it a good step up from the single Coral Edge TPU. Again, this may be considered underpowered for LLMs compared to the Hailo-8L's 13 TOPS, but it offers Still plenty of use for the right project.
Pineboard's hat AI! Dual Edge Coral TPU Bundle for Raspberry Pi 5
The Pineboards Has AI! The Dual Edge Coral TPU Bundle contains a Coral TPU but has a second M.2 socket for an additional AI module. It mounts underneath your Raspberry Pi 5 so you still have access to the GPIO header.
1 The Pineboards Has mPCIe Coral Edge TPU Bundle
Your TPU plus a side of cellular data and USB 2.0
Finally, what happens if your AI project requires internet access even though there is no WiFi? The Pineboards Hat mPCIe Coral Edge TPU Bundle is perfect for this scenario. The HAT includes a Coral Edge TPU, but offers more than just that.
This HAT also features a Nano SIM socket and an optional USB 2.0 port. This supports Wi-Fi cards and cellular modems, giving your Raspberry Pi 5 the ability to use cellular data for projects that are not in an area where you have Wi-Fi coverage.
Pineboards Has mPCIe Coral Edge TPU Bundle for Raspberry Pi 5
This PCIe HAT for the Raspberry Pi 5 includes a Coral Edge TPU for AI tasks as well as additional features. It includes an optional USB port and a Nano SIM slot, useful for various Wi-Fi cards and cellular modems.
Embark on an AI adventure with the Raspberry Pi 5
With these options for the official Raspberry Pi 5 AI Kit, you can at least dip your toes into AI projects. While you won't be able to run anything as robust as ChatGPT, once you have one of these HATs providing the NPU, there are many AI applications that the Pi is perfect for.