I fed my Home Assistant logs into a local LLM, and it found problems I'd been ignoring for months

XDA developers | March 12, 2026, 8 a.m. | Original
Now's a good a time as any to sort it out.
I consolidated 12 apps into 4 Docker containers and my homelab runs smoother than ever Qwen3.5-9B tops every AI benchmark right now, but that's not how you should pick a model