Onirix on your desktop.
Your workspace in its own window. Open models download and run on the same computer.
macOS, Windows, Linux. Free.
Ollama is running on this computer
Onirix reaches it at host.docker.internal:11434
- Qwen 2.54.7 GB
- Gemma 32.1 GB of 3.3 GB
- gpt-oss 20Babout 14 GBDownload
Open models
Two ways to bring a model.
Run it on this computer
Pick a model and click Download. The app installs Ollama, pulls the model, and serves it. Nothing leaves the machine.
Needs Onirix running on the same computer
Point at a server
Enter the address of an Ollama or OpenAI-compatible server. Onirix tests it from where it runs, then lists what it serves.
Any address your Onirix server can reach
Downloads
Pick your platform.
macOS
macOS 12 or later
- Apple siliconComing soon
- IntelComing soon
Windows
Windows 10 or later
- 64-bit installerComing soon
Linux
Any recent distribution
- AppImage, x64Coming soon
The first release is not published yet.
What it is
A client, not a server.
The app connects to an Onirix server: yours in Docker, or your team's. Accounts, documents, and permissions stay there.