Apps & Tools
eGPU Mac Mini LLM Setup
A hybrid M4 Mac Mini and RTX 3090 rig for high-speed local LLM inference
by0xSero
Is this your creation?

Description
This hardware-software hybrid setup utilizes Tinygrad's 'chestnut' to connect an NVIDIA GPU to an Apple Silicon Mac via USB-C. It leverages the Strata framework for advanced memory tiering across Mac RAM, GPU VRAM, and local storage, achieving 80 tok/s decode and over 1,000 tok/s prefill on a 2-bit quantized model.
Descriptions, tags, and model credits may be AI-generated or inferred from public sources and can be incomplete or wrong. Learn more.