
Private offline AI chat on Android with GGUF models and llama.cpp
AI & LLM · Android
An AI router and LLM gateway that keeps your API keys on your phone
Ollama Router Pro is an AI chat app and an LLM gateway in one. Each prompt is routed automatically to a fast, a balanced or a deep-reasoning tier: short questions go to quick models, harder tasks such as debugging or planning go to reasoning models. If a provider fails, the request falls back to another one.
The same app runs an OpenAI-compatible server on the phone that coding tools like Cursor, Claude Code, Windsurf, Cline or Continue can use over Wi-Fi or USB ADB. API keys live in a local vault backed by the Android Keystore with AES-256-GCM, and prompts go straight from the device to the provider. Live MCP tools let the model search the web and take actions from the chat, and a usage view shows requests, latency and routing statistics.
Only on your device, encrypted in a vault backed by the Android Keystore (AES-256-GCM). They are not uploaded to a server.
Yes. The app starts an OpenAI-compatible server on the phone, reachable over Wi-Fi or USB ADB. Tools such as Cursor, Claude Code, Cline or Continue use it as their base URL.

Private offline AI chat on Android with GGUF models and llama.cpp

Run language models on Android and serve an OpenAI-compatible API on your own Wi-Fi

Charge alarm at 80, 85, 90 or 100 percent, plus battery temperature and charge cycles
Mobile apps
I build Android apps from the first idea to the release on Google Play. Tell me briefly what your app should do.