“LLM Vision is a free, open-source Home Assistant integration that lets an AI model examine the images from your cameras and describe what is happening in those images. LLM Vision doesn’t replace systems like Frigate; it builds on top of them. Instead of just getting a notification that a person is detected, the AI model can describe what the person is doing, whether they’re carrying a package, which direction they’re going, whether they’re doing anything unusual, and even describe the person and what they’re wearing.”
In 2026, we should not still be rewinding for hours through video from cameras to spot the issue we are looking for. Imagine being able to just ask it, is the garage door open, are the trash cans outside still, when was the cat last seen, etc.
Key though is to really have a LLM running locally that does the work, saving on bandwidth, ensuring privacy, as well as for when the Internet is not reachable.
I only wish I realised all this when I bought a new mini-server and my last graphics card. If I knew how useful it was going to be, having local AI running as a service at home, I would have bought differently. Only my desktop PC has this ability, and it is not running 24/7, and it does other things.
See https://www.howtogeek.com/turn-security-cameras-into-ai-assistants-with-open-source-tool