Ensu

Ensu is Ente's local-first AI chat app. It runs a large language model (LLM) directly on your device, so your conversations stay completely private. There are no accounts required, no tracking, no usage limits, and no cost.

Think of it as a ChatGPT-like experience, except nothing ever leaves your device. It works offline, and it is free and open source.

For the full backstory on why we built Ensu and where it is headed, read the launch post: Ensu - Ente's Local LLM app.

Key highlights

• Fully private: All inference happens on your device. Your conversations are never sent to any server.
• Works offline: No internet connection needed once the model is downloaded. Perfect for flights, travel, or anywhere without connectivity.
• Image attachments: You can attach images to your messages for multimodal conversations on iOS, Android, and desktop.
• Open source: Like all Ente products, the code is fully open source and available on GitHub.
• Cross-platform: Available for iOS, Android, macOS, Linux, and Windows. There is also an experimental web version.

Availability

You can download Ensu from ente.com/ensu.

Platform	Where to get it
iOS	App Store
Android	Google Play / GitHub releases
macOS / Windows / Linux	GitHub releases
Web	ensu.ente.io (experimental)

Current status

Ensu is currently an Ente Labs project. This means we are actively iterating on the product and its direction, without committing to a fixed pricing or stability guarantee just yet. It is not as powerful as ChatGPT or Claude, but it is already useful and fun for many tasks.

Some things people have been using it for:

• Private reflection and journaling with an AI that never phones home.
• Discussing books and ideas (Ensu knows classics like the Gita and the Bible quite well, even without web search).
• Chatting on flights and in places with no internet.

Getting Help

• FAQ: Check the FAQ for common questions.
• Discord: Join the #ensu channel on Discord.
• Email: Write to [email protected] with feedback or questions.
• GitHub: Report bugs or request features on GitHub.