System Requirements

Novellis is designed to be efficient. The hardware requirements depend primarily on whether you choose to use hosted cloud models or run your own local AI.

Hosted LLM (Standard Usage)

If you are using hosted providers (e.g., OpenAI, Anthropic, Google Gemini), Novellis has very modest requirements:

• macOS: 10.15+ (Intel or Silicon)
• Windows: 10+ (64-bit)
• RAM: 8GB
• Internet: Required for AI features

Local LLM (Advanced Usage)

To run local models via Ollama or other engines entirely on your own hardware, we recommend:

macOS (Silicon Recommended)

• Processor: M1/M2/M3 (Pro/Max recommended for 7B+ models)
• RAM: 16GB unified memory (32GB+ for complex analysis)
• Storage: 500MB for application, plus storage for models (4-10GB per model)

Windows (GPU Recommended)

• GPU: NVIDIA RTX 3060+ with 8GB VRAM (minimum for local inference)
• Processor: Modern i7/i9 or Ryzen 7/9
• RAM: 16GB minimum