System Requirements
Minimum Hardware
| Component | Minimum | Recommended |
|---|---|---|
| CPU | 2 cores | 4+ cores |
| RAM | 2 GB | 8 GB+ |
| Disk | 500 MB (l3mcore) | Depends on local models |
| GPU | Not required | Optional (for local ONNX/GGUF) |
Raspberry Pi
l3mcore is so lightweight that it can run on a Raspberry Pi 4 with 4 GB of RAM. Ideal for homelabs.
Software
- Python: 3.9 or higher
- pip: included with Python
- git: to clone the repository
- Ollama (optional): if you want to use local backends with Ollama
Python Dependencies
The main dependencies are installed automatically with setup.sh:
flask
sentence-transformers # for ML embeddings
rapidfuzz # for keyword fallback
litellm # for external APIs (OpenAI, Anthropic, etc.)
onnxruntime # for local ONNX models (optional)
OS Compatibility
| Operating System | Status |
|---|---|
| Linux (Ubuntu 20.04+, Debian, Arch) | Supported |
| macOS (12+) | Supported |
| Windows (WSL2) | Supported via WSL2 |
| Native Windows | Not tested |
| Raspberry Pi OS | Supported |