Installation
Lumen is a desktop app for macOS. Installing it takes a minute; connecting your first model takes one more.
Requirements
| What | Minimum |
|---|---|
| System | macOS 14 Sonoma or later, Apple Silicon or Intel. Windows is coming soon. |
| Models | An API key for at least one provider |
| Projects | Any local folder. Git is optional but recommended. |
Install the app
- DownloadGet the disk image from the download page. One build runs on Apple Silicon and Intel.
- Move to ApplicationsOpen the disk image and drag Lumen into Applications.
- Open LumenThis early build isn't notarized by Apple yet, so on first launch macOS says it can't check the app. Open System Settings → Privacy & Security, scroll down and click Open Anyway. You only do this once.
Connect a model provider
On first launch Lumen asks for a key. Later, open Settings (⌘,) and paste an API key for each provider you want to use, then click Check. Keys are stored in the macOS Keychain and never leave your machine except to call the provider.
| Model | Where to get a key |
|---|---|
| Anthropic Console | |
| OpenAI Platform | |
| Google AI Studio | |
| DeepSeek Platform | |
| Alibaba Cloud Model Studio | |
| Mistral La Plateforme | |
| xAI Console | |
No key: install Ollama and run ollama pull llama3.2 |
Tip
You don't need every provider. With one key, that model works alone as both manager and specialist. Add keys from more providers to split the work across the crew.
Next
Head to the Quickstart to create your first workspace and crew.