Sunday, January 28, 2024

NOW! Raspberry Pi 5 can run open-source AI models

 Raspberry Pi has been out for a few months now, and we've had the opportunity to see what it can do. Now, a maker and developer named Data Slayer has explored how the Pi 5 can run advanced AI large language models (LLMs) in his latest video. In the video, we can witness the impressive capabilities of the Pi 5 in utilizing local AI models.

Data Slayer uses a tool called Olama to handle and switch between LLMs. This is done through the command line and doesn't require internet access as everything is stored on the Pi 5. In the video, Data Slayer provides a detailed look at what various models can accomplish when operated on the Pi 5.

The first LLM he showcases is called LLaVa, used for image analysis. For a test, Data Slayer takes a selfie and passes it through the Pi to see what LLaVa will detect. True to form, it accurately identifies the subject of the image, indicating that it's a picture of a man wearing a hoodie in a building in an urban environment.

The next example is a model called Llama 2, which handles entertaining text-based questions. From the start, he asks for a spicy mayonnaise recipe and proceeds to test the model with a bit of historical facts. To conclude his experiment, he asks the model to generate some basic Linux commands.

This is a really cool example to see what you can do with Raspberry Pi 5 without the need for additional hardware. Nevertheless, you can enhance the experience by adding external storage or a TPU for better performance. In the video created by Data Slayer, he uses a Raspberry Pi 5 with 8GB of RAM and a 256GB microSD card. If you want to explore this Raspberry Pi project in more detail or just want to learn more about how everything works, check out the original video on YouTube by Data Slayer and be sure to follow him for future projects.



Post a Comment