OLABHome AI Cluster
OLAB - the O comes from Oljasz - is my personal infrastructure lab and the backbone for everything else on this page. A distributed AI inference cluster built from 4x Mac Mini M4 machines connected via Thunderbolt mesh, paired with a high-end NVIDIA GPU server for heavy compute. Full monitoring stack with Grafana, Prometheus, and Langfuse. Security pipeline with LLM Guard and Llama Guard 3. Auto-sleep with wake-on-request. All services, CI pipelines, and staging environments for FutureGrid and WordHunt run on this infrastructure.
- exo
- ollama
- docker
- grafana
- langfuse
- thunderbolt
- nvidia