Local and Native AI.
Uncompromising Privacy.

FolliA is a native Android client for LLMs, designed to interact with your own AI models. Fast, local, and secure: your data stays where it belongs.

๐Ÿ›ก๏ธ

Total Sovereignty

Connect your API keys or your own Ollama server via VPN. Zero intermediates, no tracking, absolute privacy.

๐ŸŽ™๏ธ

Hands-Free Voice Mode

Chat naturally with your AI using optimized audio logic that prevents echo feedback and infinite mic loops.

๐Ÿ”„

Background Persistence

No more generation cut-offs. Switch between apps freely while inference continues to process smoothly in the background.

โ˜๏ธ

Nextcloud Synchronization

Securely backup and sync your entire conversation history independently using native WebDAV integration.

๐Ÿ“ฑ

Adaptive Tablet UI

Experience a dual-pane layout and navigation rail tailored for large screens, instead of just a stretched mobile layout.

๐Ÿš€

Independent Updates

Stay up to date without relying on centralized app stores with an autonomous pipeline fetching GitHub Releases.

Architecture Breakdown

Private Inference

Run your models on your own hardware. The heavy cryptographic lifting stays entirely at home.

Secure Tunnel

FolliA communicates directly and securely with your environment via local network or your VPN setup.

Zero Proxies

No external cloud servers, no hidden telemetry. Your prompts and answers belong to you exclusively.