Transform your iPhone or iPad into a private AI inference server. Run large language models entirely on-device, offering an OpenAI-compatible API to your local network. Chat with your model from any browser without cloud reliance or subscriptions.
2.0.4
July 8, 2026
- Reliability improvements - Added Apple Intelligence model - Added Anthropic API support
MoreThe vision model feature is still being finalized and will land in the next update, coming very soon. If you bought the app specifically for this, email me at contact@linosec.com and I'll make sure you're looked after. Thanks for your patience.
LLM Server turns your iPhone or iPad into a private AI inference server. It allows you to run large language models entirely on-device and exposes an OpenAI-compatible API to your local network.
LLM Server is a paid application, priced at $6.99. There are no subscriptions or in-app purchases mentioned.
No, LLM Server does not contain advertisements. It is a paid application designed for a seamless, ad-free user experience.
LLM Server is supported on iPhones and iPads. The app is rated for users aged 17 and older.
Yes, LLM Server is designed for fully offline operation. Once a GGUF model is downloaded, inference runs locally on your device's Apple Metal GPU without requiring an internet connection.
LLM Server supports any GGUF model from Hugging Face, including LLaMA, Mistral, Phi, Gemma, Qwen, and DeepSeek, as well as other llama.cpp-supported architectures.
The latest version of LLM Server is 1.4, last updated on June 29, 2026. The release date was May 1, 2026.
LLM Server provides enterprise-grade security with TLS/HTTPS encryption, API key authentication, and bind control to manage network access.
