Quick Start

The easiest method is to run the setup script that sets up everything for you.

Option 1 — Run directly

Copy and paste this into your terminal:

bash -c "$(curl -fsSL https://raw.githubusercontent.com/rosmur/local-llm-setup/main/setup-local-llm.sh)"

Option 2 — Download first

Download the script from github.com/rosmur/local-llm-setup, then run it:

bash Downloads/setup-local-llm.sh

Script Notes

• The script is written to be readable and to ask before it does anything.
• Every step explains what it is about to do, shows the exact command, and waits for you to type y. Typing anything else skips that step. Ctrl-C quits at any point.
• Re-running the script is always safe. Every step detects work that is already done and skips it. An interrupted run — a dropped connection, a closed laptop lid — is recovered by just starting again. The menu marks which models you already have, and nothing is downloaded twice.

High Level Steps

1. llama.cpp — Installed via Homebrew (macOS) or the official direct installer (macOS, Linux, Windows)
2. Model — You pick from a menu of open-weight models, or type 0 to skip and download later
3. pi — The coding assistant is installed
4. pi-llama plugin — Installed so pi auto-discovers your local model
5. llama serve — You run the model server; no config files, no launcher script

After the script finishes

See the Usage page for how to start the engine and use pi.