New Update Makes AI Run Faster on Everyday Intel Laptops

Most people use AI through websites like ChatGPT, which means your data goes to a company's servers. But there's another way: running AI directly on your own computer. A free, open-source program called llama.cpp makes this possible. This latest update — version b10723 — makes that local AI significantly faster on Intel's built-in graphics chips, often called Intel Xe-LP GPUs. These chips are common in mainstream laptops, not just expensive gaming machines.

So what changed? The update refines how the software performs math calculations for AI models. Specifically, it optimizes the way data is packed and processed for Intel's graphics hardware. Think of it like reorganizing a warehouse so workers can grab boxes faster. The result is that generating responses (called "token generation" in technical terms) speeds up, and the computer uses its hardware more efficiently.

Why should you care? If you use AI on a laptop with Intel integrated graphics, you'll likely notice faster replies, especially when asking for longer responses. It also means running AI locally becomes more practical — you can get helpful answers even without an internet connection, and your personal questions never leave your device. That's a big deal for privacy-conscious people or anyone working with sensitive information.

The catch: these improvements are tuned specifically for Intel Xe-LP graphics. If you have an older Intel chip, an AMD computer, or a Mac, you won't benefit directly. Also, running AI on a laptop is still slower than using a big cloud server — but each update closes that gap, making local AI more attractive for everyday use.