I gave a local LLM access to my Docker containers, and it replaced my monitoring scripts
… All the computing was done in my gaming PC that had an RTX 4070 Ti. My PC was on almost all the time yes, I pay a lot for the electricity bill , so hosting the local model on my PC was the right option for me. After that, it was a straightforward process. Gemma 4 running locally via Ollama on my PC. …