Worked on the vllm-project/llm-compressor repository to deliver a complete quantization example for the InternVL3-8B-hf model, focusing on enabling reproducible workflows and practical evaluation pipelines for cost-effective deployment. The contribution included all steps from model loading and dataset preparation to preprocessing and evaluation, with comprehensive documentation and a detailed testing plan to ensure verification and reusability. The work was implemented in Python and leveraged skills in data processing, machine learning, and model optimization. All changes were isolated to the quantization example, providing a reusable template for future models and aligning with collaborative code quality standards.