Guides for running LLMs on your phone
Practical, step-by-step articles on running models on Android, using your phone as an OpenAI-compatible endpoint and connecting your editor.
How to Run an LLM on Android with Local GGUF Models
Running a language model on your phone is practical once you understand GGUF files, quantization and what still needs a network. This guide walks through the concepts and then shows how to start the on-device server in Ollama Local AI.
Read the guideUse Cursor with a Local LLM on Your Phone
Cursor can send its OpenAI-style requests to any compatible base URL, including the one an Android phone serves on your Wi-Fi. This guide covers the exact settings, how to find the endpoint and model id, and what to check when the connection fails.
Read the guideUse Your Android Phone as a Local LLM in VS Code
Continue, Cline and Roo Code all accept an OpenAI-compatible provider, and Windsurf can point custom OpenAI model endpoints at a LAN address. This guide shows the values to enter so those tools talk to a model served from your phone.
Read the guideOpenAI-Compatible API on Android: Endpoints, Base URL and Code Examples
Your Android phone can answer the same requests you send to a hosted chat API, so existing scripts and tools work against it by changing a base URL. This guide covers the endpoints, the URL format, working code, and how to keep the server on your own network.
Read the guideOllama Local AI vs Ollama: Two Different Things With a Shared Word
Ollama is an open-source tool for running models on a computer. Ollama Local AI is an independent Android app that is not affiliated with it. Here is how they differ, how they can work together, and which one fits your situation.
Read the guide