Get started with Monadd-AI, the offline LLM inference engine that enables ultra-large SOTA MoE models on consumer-grade hardware.
Current Version: Unreleased Alpha 0.1 (target Public-Beta release Q4 2025)
Monadd-AI will be made available for Windows, macOS, and Linux. Once released, select your preferred platform to get started.
Check if your system meets the minimum requirements to run Monadd-AI efficiently.
| Component | Minimum | Recommended |
|---|---|---|
| Operating System |
Windows 10 (64-bit) macOS 12 Monterey Ubuntu 20.04 LTS |
Windows 11 (64-bit) macOS 13 Ventura or newer Ubuntu 22.04 LTS or newer |
| RAM | 16 GB | 32 GB or more |
| GPU | e.g. NVIDIA GPU with 16 GB VRAM (CUDA 11.8+) | NVIDIA or AMD GPU with 24 GB VRAM or more (e.g., RTX 4090 RTX A/Quadro, AMD Radeon RX 7900 XTX) |