Skip to content

Animation: eight low-poly animals, each with its own personality, chat around a question put to Lucania. They show its features: thinking mode, streaming and Markdown rendering, a separate memory for each model, folders and pinning, several Ollama servers including a remote one over WireGuard with remote shutdown, 13 themes, English and French. Everything stays on your computer.

Lucania – Local chat for your Ollama models

The whole pack runs on your machine.

A local chat interface, Claude.ai-style, for your Ollama models.

Windows 10/11 64-bit · free · open source (MIT)

  • So where do my conversations go?
  • Wolf · local guardian

    Nowhere. Lucania talks to Ollama on your machine and keeps everything local. Nothing goes to the cloud.

Preview

Your models, at home.

Here is the interface in action. Open a conversation, ask a question, turn thinking on, pick a model or a theme.

Interactive demonstration of the Lucania interface

Understanding Ollama

How does Ollama run a model on my computer?
The question is about local execution. I will start by explaining that the model runs on the machine, then remind what Lucania adds, and finish with the command to download the model.

Ollama runs the model directly on your machine: nothing leaves your computer. Lucania connects to it and brings you:

  • Streaming: answers appear as they are written.
  • Markdown: lists, code blocks and formatting are rendered.
  • Memory: each model keeps its own, locally.

To get the model used in this example:

ollama pull llama3.2
ollama run llama3.2
llama3.2:3b · 2:32 PM

Get started

Ready to chat locally?

Download Lucania, plug in Ollama and keep your conversations on your machine.

Windows 10/11 64-bit · free · open source (MIT)