Haoyang Li

Haoyang Li is a Software Engineer on the Model Quantization team at AMD and one of the main contributors to AMD-Quark. His research interests include model quantization techniques, inference framework optimization, and high-performance computing.

Posts by Haoyang Li

October 29, 2025

High-Accuracy MXFP4, MXFP6, and Mixed-Precision Models on AMD GPUs

Learn to leverage AMD Quark for efficient MXFP4/MXFP6 quantization on AMD Instinct accelerators with high accuracy retention.

https://rocm.blogs.amd.com/software-tools-optimization/mxfp4-mxfp6-quantization/README.html

August 26, 2025

QuickReduce: Up to 3x Faster All-reduce for vLLM and SGLang

Quick Reduce speeds up LLM inference on AMD Instinct™ MI300X GPUs with inline-compressed all-reduce, cutting comms overhead by up to 3×

https://rocm.blogs.amd.com/artificial-intelligence/quick-reduce/README.html