Chart standing · US
About
Transform your iPhone, iPad, or Mac into a private, local LLM AI server. This powerful utility allows you to run Large Language Models (LLMs) powered by Apple Intelligence directly on your device, providing an OpenAI-compatible API and Ollama compatible API for your development and testing needs. Run a local AI LLM server on your Apple Intelligence enabled device to give private, local edge AI abilities to your applications. Note: This app requires iOS 26.0/macOS 26.0 or above to run Apple Intelligence models. **Key Features:** * **Local Server Hosting:** Run a fully functional LLM server right from your pocket. No internet connection is required. * **OpenAI-Compatible API:** Seamlessly integrate with existing tools and clients that support standard chat completion endpoints (OpenAI and Ollama API). * **Real-Time Monitoring:** Keep track of your server's information with a live status dashboard showing IP address, active port, and logs. * **Live Logs:** View real-time server logs to debug requests and monitor performance instantly. * **Developer Friendly:** Includes built-in curl command example to quickly test your server. Whether you are an AI researcher, a developer building private apps, or an enthusiast exploring local intelligence, this app provides the local LLM AI infrastructure you need using the devices you already have. This app is not affiliated with nor is endorsed by Apple. Use responsibly and in accordance with Apple's terms of service.
What's new · 1.5
• New /api/generate endpoint — the Ollama single-prompt API that many Python and Ollama examples use. • More reliable during long, looped sessions: a stuck request can no longer wedge the server and stop it from answering. • The on-device model's context-window size is now shown in Server Info and reported via /api/show. • Fixes and refinements.
Details