Free Open-Source AI Update Lets Your Devices See and Hear
Your gadgets can now run AI that understands images and sound — privately, offline.
A popular open-source project called llama.cpp released a new update, version b10580. The big addition is support for a new AI model called dots3-note that can handle two things at once: vision (understanding images) and audio (understanding sound). In plain terms, this update lets software on your own computer or phone process pictures and speech, instead of relying on big cloud servers.
Why does this matter? Most AI tools you use, like chatbots or photo assistants, send your data to a company's servers. That can raise privacy concerns and requires an internet connection. llama.cpp is different — it's software designed to run AI models locally on your own device. With this update, a future app on your laptop or phone could identify objects in a photo or transcribe a conversation entirely on-device, with no data leaving your machine.
The catch: this is a pre-release for developers, not a polished app you can download today. There are many versions for different systems — Mac, Windows, Linux, Android, and more — which shows how flexible it is, but also how technical it is. You'd need some comfort with installation and command-line tools to try it. Expect bugs and incomplete features.
Still, this is a clear step toward AI that's private, free, and works offline. As projects like this mature, everyday people may soon get apps that see and hear without paying a subscription or sacrificing privacy.
- llama.cpp, a free open-source AI tool, now supports a model that understands both images and sound.
- This runs locally on your own computer or phone, so your data doesn't have to go to the cloud.
- It's an early pre-release for developers — not yet a simple app for everyday users.
Why It Matters
Private, offline AI that understands images and sound is getting closer to your everyday devices.