- Posted on
- Featured Image
A practical Linux guide to running local LLMs with Ollama: why it’s private, fast, and scriptable; how to install on Debian/Ubuntu, Fedora/RHEL, and openSUSE (or Docker); run and manage models like Llama 3; call via Bash and the HTTP API; configure systemd, ports, and storage; use CPU/GPU; build a custom model with a Modelfile; and troubleshoot performance, disk space, and networking.