Ollama Local AI

Ollama Local AI is the ultimate companion for developers looking to integrate local Large Language Models (LLMs) seamlessly with their favorite IDE development workflows. Turn your Android device into a secure, ultra-fast, and switchable local-network OpenAI-compatible LLM proxy.

Point Cursor, VS Code, Antigravity, Windsurf, or any modern coding assistant directly to your phone's LAN IP address and start using local models or API keys without configuring multiple endpoints on your development machine.

### Key Features:
- **Zero-Config Local Hosting**: Spins up a highly optimized embedded HTTP server (NanoHTTPD) right on your Android phone.
- **OpenAI-Compatible API**: Implements standard endpoints (/v1/chat/completions, /v1/models, /health) allowing drop-in compatibility.
- **Instant Provider Switching**: Save multiple configurations (Ollama local inference, NVIDIA, OpenAI, Hugging Face, custom endpoints) and switch between them dynamically in one click without altering your IDE config.
- **Real-Time Traffic Visualizer**: Watch request flow, packet routing, and performance stats live using an interactive canvas visualizer.
- **Lazy Token Bucket Rate Limiting**: Built with a battery-efficient token-bucket pruner to prevent API overuse with zero background CPU/battery drain.
- **Secure Encrypted Storage**: All sensitive API keys and provider credentials are encrypted using Android's EncryptedSharedPreferences (AES-256).
- **Offloaded I/O Operations**: Fluid Compose UI that offloads network routing, preference writes, and JSON loading to background threads (Coroutines + Dispatchers.IO) for a stutter-free experience.

### Why choose Ollama Local AI?
Eliminate the complexity of managing API keys, proxies, and configurations across multiple machines. Maintain absolute control over your keys and models while keeping your development environments light, fast, and unified.

Optimize your LLM development today with Ollama Local AI!