3 posts tagged with “ollama”
How I stood up a fully self-hosted AI chat stack with retrieval over my own homelab documentation, and the gotchas that came with wiring RAG together on a single-GPU box.
How I built a workflow that uses a local LLM to analyze Zabbix alerts and send intelligent Telegram notifications - no cloud APIs, no subscription fees.
How I configured NVIDIA GPU passthrough for Plex hardware transcoding and Ollama LLM inference in privileged Proxmox LXC containers, including driver gotchas and VRAM limitations.