2025年2月14日 · A Raspberry Pi Zero W has been modified to function as a USB stick running a local large language model (LLM) entirely offline. This innovative device, created by Binh Pham from Build With Binh, interacts with users through USB mass storage, allowing AI-generated content to be written dynamically by simply creating a file.

2025年2月20日 · Youtuber and tech enthusiast Binh Pham has recently built a portable plug-and-play AI and LLM device housed in a USB stick called the LLMStick and built around a Raspberry Pi Zero W. This device portrays the concept of a local plug-and-play LLM which you can use without the internet.

2025年2月17日 · By setting up the USB device as a composite device which presents a filesystem to the host computer, all a user has to do to interact with the LLM is to create an empty text file with a filename ...

2025年2月17日 · To use the LLM, all you need to do is plug in the USB stick, create an empty text file, and give it a name. The model automatically generates text and saves it to the file. While it’s not as...

2025年2月17日 · Inside the 3D-printed shell of this somewhat-oversized USB stick is a Raspberry Pi Zero single-board computer and a shield that adds a male USB port for interfacing with a host computer. Much of the work of getting an LLM to run on this platform has already been handled by the llama.cpp project, so Pham leveraged that.

2025年2月17日 · Binh Pham experimented with a Raspberry Pi Zero, effectively turning the device into a small USB drive that can run an LLM locally with no extras needed. The project was largely facilitated...

2025年3月3日 · The major PC and device maker indicates that the AI Stick will deliver AI features, such as local Large Language Models (LLM) and AI-enhanced graphics apps, on the devices it connects with....

2025年2月20日 · YouTuber et le passionné de technologie Binh Pham a récemment construit un appareil AI et LLM portables Plug-and-Play Institué dans un bâton USB appelé LLMStick et construit autour d’un Raspberry Pi Zero W. Cet appareil décrit le concept d’un jeu de fiche local LLM que vous pouvez utiliser sans Internet.

2025年2月17日 · Successfully runs three different LLM models: Tiny Stories (15M parameters), Lamini (77M parameters), Smollm (135M parameters) The smallest model generates tokens at ~200ms per token, while larger models take about 2.5 seconds per token. When you create a new file on the USB drive, it detects the filename and automatically generates content!

💪 I created the first USB in the world that has LLM running locally on it.Cherry on top of the cake, it requires no dependency, you can connect it to any co...