Free Software That Runs AI Chatbots on Your Laptop Just Got an Update

llama.cpp is a free, open-source program that lets you run AI chatbots on your own computer instead of sending every question to a company's servers. Think of it as the difference between baking bread at home and ordering it from a bakery every day — same loaf, but nobody else sees your order. Its latest release, tagged b11149, was published on 23 September and is a "pre-release," meaning it's a snapshot for testers rather than a polished final version.

The actual change in this release is tiny and won't affect how you use it. The developers added a small option to their testing tools that lets them check how the software performs on one specific type of computer hardware at a time. In plain terms: it's a new knob for quality control, not a new feature. If you already use llama.cpp, nothing about your experience changes today.

What's genuinely interesting is the list attached to every release: the sheer range of devices this free software now runs on. That includes Apple's Mac chips, ordinary Windows and Linux PCs, Android phones, phones using Qualcomm's Snapdragon chips (with help from the graphics and AI chips inside them), and even Chinese-made Huawei server chips. There are separate builds for Nvidia, AMD, and Intel graphics cards too. In other words, the same free tool works on roughly whatever you already own.

The real-world payoff is privacy and money. Running AI locally means your documents, emails, and questions never leave your device. There's no monthly fee, and it works on a plane with no Wi-Fi. The catch is that your computer's speed and memory set the ceiling: small models run fine on a normal laptop, while the smartest ones still need an expensive machine.