This path is for you if you want Georgi Gerganov's llama.cpp engine: a local run, a GGUF file, and a small API on your own machine. No part is live on this page yet. Start with Run open models, which walks a desktop runner first, and come back when part 1 is up.
- 1 What is llama.cpp? A free program that runs AI models on your computer Treat llama.cpp as a building block, not a software vendor. It powers Ollama and LM Studio behind the scenes. Compile once, run offline, no per-seat bill. Scheduled · November 13, 2026
- 2 The easiest way to install llama.cpp, and when to stop and use an app instead Prefer package managers or official release zips for llama.cpp. Compile from source only when CUDA on Linux or bleeding-edge models force it. Stop when tokens outpace reading. Scheduled · November 19, 2026
- 3 How to run your first AI model with llama.cpp Load a GGUF file, set the model path, temperature and GPU layers, and read llama.cpp's timing line. Current llama-cli chats by default; add -st for a single answer. Scheduled · November 24, 2026
- 4 How to make llama.cpp run faster on your computer Diagnose llama.cpp slowness with KV cache size, physical-core thread counts, and memory bandwidth limits. Quantize the KV cache before you buy new silicon. Scheduled · November 28, 2026
- 5 How to share an AI model on your computer with other programs using llama-server Run llama-server with host, port, parallel slots, and cont-batching. Point the official OpenAI client at localhost. Keep long system prompts warm with prefix cache. Scheduled · December 1, 2026
- 6 How to add a ChatGPT-style chat window to llama.cpp with Open WebUI Put a real UI on llama-server: Open WebUI for teams, Continue for editors, desktop apps for zero Docker. Use host.docker.internal, not localhost, from containers. Scheduled · December 4, 2026
- 7 How to fix the most common llama.cpp problems Triage llama.cpp fails: read stderr, check GGUF magic (curl -L), match CPU flags, shrink offload or quantize KV, fix chat templates. Add a self-healing wrapper for edge boxes. Scheduled · December 6, 2026