Ente Labs has launched Ensu, an offline LLM application that runs entirely on users' devices without cloud dependency. The iOS, Android, macOS, Linux, and Windows app provides ChatGPT-like functionality while maintaining complete privacy and zero cost. Built with Rust core logic and native or Tauri interfaces, Ensu supports image attachments and includes experimental web versions.

The project stems from Ente's philosophy that LLMs are too important to be left to big tech. Drawing parallels to their successful Ente Photos platform, which brought face recognition and natural language image search to local devices, the team believes smaller models are approaching a capability threshold where they'll provide sufficient utility for most users. The app currently lacks web search but performs well with classic texts like the Gita and Bible.

While not yet matching ChatGPT or Claude Code in power, Ensu offers practical uses including private thought processing and offline conversations during flights. The team plans to add optional encrypted sync via Ente accounts, though this feature was delayed to refine the persistence architecture. As an Ente Labs project, Ensu remains in early development without pricing or stability guarantees. The team actively seeks user feedback to shape future directions, which could range from specialized interfaces to agent-like functionality.