Hi Janusz Nowak,
What if you could point a question at a live stream and get an answer back?
"Is anyone in this frame without a helmet?"
"Is there smoke or fire visible?"
"What text is on screen right now?"
That's the Ollama Vision AI Plugin for Ant Media Server — real-time video understanding powered by local vision models running through Ollama. It samples frames from your live streams, asks the model, and stores every answer and alert for you to pull over REST.
The catch most teams care about: it all runs on your hardware. Nothing leaves your infrastructure, which means lower latency, no cloud AI costs, and privacy by design.
It's the first in a new series of plugins we're rolling out, and we're inviting a small group into the private beta first.
