Brian Dellabetta

The LLM Compressor 0.8.0 release introduces quantization workflow enhancements, extended support for Qwen3 models, and improved accuracy recovery.

LLM Compressor 0.7.0 brings Hadamard transforms for better accuracy, mixed-precision FP4/FP8, and calibration-free block quantization for efficient compression.

In this recording, we demonstrate how to compose model compression experiments, highlighting the benefits of advanced algorithms requiring custom data sets and how evaluation results and model artifacts can be shared with stakeholders.

Optimize model inference and reduce costs with model compression techniques like quantization and pruning with LLM Compressor on Red Hat OpenShift AI.