Description

A high-performance inference stack designed to run large models like GLM-5.3-Flash on single consumer GPUs. It utilizes a novel approach offloading model parts to NVMe and DDR4 RAM while leveraging the GPU for computation, using exl3-3bpw quantization to maintain performance.

Descriptions, tags, and model credits may be AI-generated or inferred from public sources and can be incomplete or wrong. Learn more.