Raspberry Pi

1908 Articles

A Trail Camera Built With Raspberry Pi

September 30, 2025 by 9 Comments

You can get all kinds of great wildlife footage if you trek out into the woods with a camera, but it can be tough to stay awake all night. However, this is a task you can readily automate, as [Luke] did with his DIY trail camera.

A Raspberry Pi Zero 2W serves as the heart of the build. It’s compact and runs on very little power, but also provides a good amount more processing power than the original Raspberry Pi Zero. It’s kitted out with the Raspberry Pi AI Camera, which uses the Sony IMX500 Intelligent Vision Sensor — providing a great platform for neural networks doing image classification and similar machine learning tasks. A Witty Pi power management module is used both for its real time clock and to schedule start-ups and shutdowns to best manage the power on offer from the batteries. All these components are wrapped up in a 3D printed housing to keep the Pi safe out in the wild.

We’ve seen some neat projects in this vein before.

Continue reading “A Trail Camera Built With Raspberry Pi” →

Two hands working a TekaSketch

TekaSketch: Where Etch A Sketch Meets Graph Theory

September 29, 2025 by 6 Comments

The Etch A Sketch was never supposed to meet a Raspberry Pi, a camera, or a mathematical algorithm, but here we are. [Tekavou]’s Teka-Cam and TekaSketch are a two-part hack that transforms real photos into quite stunning, line-drawn Etch A Sketch art. Where turning the knobs only results in wobbly doodles, this machine plots out every curve and contour better than your fingertips ever could.

Essentially, this is a software hack mixed with hardware: an RPi Zero W 2, a camera module, Inkplate 6, and rotary encoders. Snap a picture, and the image is conveyed to a Mac Mini M4 Pro, where Python takes over. It’s stripped to black and white, and the software creates a skeleton of all black areas. It identifies corner bridges, and unleashes a modified Chinese Postman Algorithm to stitch everything into one continuous SVG path. That file then drives the encoders, producing a drawing that looks like a human with infinite patience and zero caffeine jitters. Originally, the RPi did all the work, but it was getting too slow so the Mac was brought in.

It’s graph theory turned to art, playful and serious at the same time, and it delivers quite unique pieces. [Tekavou] is planning on improving with video support. A bit of love for his efforts might accellerate his endeavours. Let us know in the comments below!

Continue reading “TekaSketch: Where Etch A Sketch Meets Graph Theory” →

Fully-Local AI Agent Runs On Raspberry Pi, With A Little Patience

September 28, 2025 by 25 Comments

[Simone]’s AI assistant, dubbed Max Headbox, is a wakeword-triggered local AI agent capable of following instructions and doing simple tasks. It’s an experiment in many ways, but also a great demonstration not only of what is possible with the kinds of open tools and hardware available to a modern hobbyist, but also a reminder of just how far some of these software tools have come in only a few short years.

Max Headbox is not just a local large language model (LLM) running on Pi hardware; the model is able to make tool calls in a loop, chaining them together to complete tasks. This means the system can break down a spoken instruction (for example, “find the weather report for today and email it to me”) into a series of steps to complete, utilizing software tools as needed throughout the process until the task is finished.

Continue reading “Fully-Local AI Agent Runs On Raspberry Pi, With A Little Patience” →

The New Raspberry Pi 500+: Better Gaming With Less Soldering Required

September 25, 2025 by 36 Comments

When Raspberry Pi released the Pi 500, as essentially an RPi 5 integrated into a chiclet keyboard, there were rumors based on the empty spots on the PCB that a better version would be released soon. This turned out to be the case, with [Jeff Geerling] now taking the new RPi 500+ to bits for some experimentation and keyboard modding.

The 500’s case was not designed to be opened, but if you did, you’d find that there was space allocated for a Power-Over-Ethernet section as well as an M.2 slot, albeit with all of the footprints unpopulated. Some hacking later and enterprising folk found that soldering the appropriate parts on the PCB does in fact enable a working M.2 slot. What the 500+ thus does is basically do that soldering work for you, while sadly not offering a PoE feature yet without some DIY soldering.

Perhaps the most obvious change is the keyboard, which now uses short-travel mechanical switches – with RGB – inside an enclosure that is now fortunately easy to open, as you may want to put in a different NVMe drive at some point. Or, if you’re someone like [Jeff] you want to use this slot to install an M.2 to Oculink adapter for some external GPU action.

After some struggling with eGPU devices an AMD RX 7900 XT was put into action, with the AMD GPU drivers posing no challenge after a kernel recompile. Other than the Oculink cable preventing the case from closing and also losing the M.2 NVMe SSD option, it was a pretty useful mod to get some real gaming and LLM action going.

With the additions of a presoldered M.2 slot and a nicer keyboard, as well as 16 GB RAM, you have to decide whether the $200 asking price is worth it over the $90 RPi 500. In the case of [Jeff] his kids will have to make do with the RPi 500 for the foreseeable future, and the RPi 400 still finds regular use around his studio.

Continue reading “The New Raspberry Pi 500+: Better Gaming With Less Soldering Required” →

Moondream title with man's face visible in background.

Using Moondream AI To Make Your Pi “See” Like A Human

September 23, 2025 by 9 Comments

[Jaryd] from Core Electronics shows us human-like computer vision with Moondream on the Pi 5.

Using the Moondream visual language model, which runs directly on your Raspberry Pi, and not in the cloud, you can answer questions such as “are the clothes on the line?”, “is there a package on the porch?”, “did I leave the fridge open?”, or “is the dog on the bed?” [Jaryd] compares Moondream to an alternative visual AI system, You Only Look Once (YOLO).

Processing a question with Moondream on your Pi can take anywhere from just a few moments to 90 seconds, depending on the model used and the nature of the question. Moondream comes in two varieties, based on size, one is two billion parameters and the other five hundred million parameters. The larger model is more capable and more accurate, but it has a longer processing time — the fastest possible response time coming in at about 22 to 25 seconds. The smaller model is faster, about 8 to 10 seconds, but as you might expect its results are not as good. Indeed, [Jaryd] says the answers can be infuriatingly bad.

In the write-up, [Jaryd] runs you through how to use Moonbeam on your Pi 5 and the video (embedded below) shows it in action. Fair warning though, Moondream is quite RAM intensive so you will need at least 8 GB of memory in your Pi if you want to play along.

If you’re interested in machine vision you might also like to check out Machine Vision Automates Trainspotting With Unique Full-Length Portraits.

Continue reading “Using Moondream AI To Make Your Pi “See” Like A Human” →

A black, rectangular box is shown, with a number of waterproof screw connectors on the front.

A Ruggedized Raspberry Pi For Sailors

September 20, 2025 by 33 Comments

Nautical navigation has a long history of innovation, from the compass and chronometer to today’s computer-driven autopilot systems. That said, the poor compatibility of electronics with saltwater has consequently created a need for rugged, waterproof computers, a category to which [Matti Airas] of Hat Labs has contributed with the open-source HALPI2.

Powered by the Raspberry Pi Compute Module 5, the electronics are housed in a heavy duty enclosure made of aluminium, which also serves as a heat sink, and closes with a waterproof seal. It has a wide variety of external connectors, all likewise waterproofed: power, HDMI, NMEA 2000 and NMEA 0183, Ethernet, two USB 3.0 ports, and an external WiFi or Bluetooth antenna. The external ports are plugged into the carrier board by short extension cables, and there are even more ports on the carrier board, including two HDMI connectors, two MIPI connectors, four USB ports, and a full GPIO header. The case has plugs to install additional PG7 or SP13 waterproof connectors, so if the existing external connectors aren’t enough, you can add your own.

Besides physical ruggedness, the design is also resistant to electrical damage. It can run on power in the 10-32 volt range, and is protected by a fuse. A supercapacitor bank preserves operation during a power glitch, and if the outage lasts for more than five seconds, can keep the system powered for 30-60 seconds while the operating system shuts down safely. The HALPI2 can also accept power over NMEA 2000, in which case it has the option to limit current draw to 0.9 amps.

The design was originally created to handle navigation, data logging, and other boating tasks, so it’s been configured for and tested with OpenPlotter. Its potential uses are broader than that, however, and it’s also been tested with Raspberry Pi OS for more general projects. Reading through its website, the most striking thing is how thoroughly this is documented: the site describes everything from the LED status indicators to the screws that close the housing – even a template for drilling mounting holes.

Given the quality of this project, it probably won’t surprise you to hear this isn’t [Matti]’s first piece of nautical electronics, having previously made Sailor HATs for the ESP32 and the Raspberry Pi.

Regretfully: $3,000 Worth Of Raspberry Pi Boards

September 20, 2025 by 53 Comments

We feel for [Jeff Geerling]. He spent a lot of effort building an AI cluster out of Raspberry PI boards and $3,000 later, he’s a bit regretful. As you can see in the video below, it is a neat build. As Jeff points out, it is relatively low power and dense. But dollar for dollar, it isn’t much of a supercomputer.

Of course, the most obvious thing is that there’s plenty of CPU, but no GPU. We can sympathize, too, with the fact that he had to strip it down twice and rebuild it for a total of three rebuilds. One time, he decided to homogenize the SSDs for each board. The second time was to affix the heatsinks. It is always something.

With ten “blades” — otherwise known as compute modules — the plucky little computer turned in about 325 gigaflops on tests. That sounds pretty good, but a Framework Desktop x4 manages 1,180 gigaflops. What’s more is that the Framework turned out cheaper per gigaflop, too. Each dollar bought about 110 megaflops for the Pis, but about 140 for the Framework.

Continue reading “Regretfully: $3,000 Worth Of Raspberry Pi Boards” →