In the realm of high-definition video content, video upscaling technologies have become crucial for enhancing resolution and preserving visual details. This paper introduces an extensive benchmarking platform specifically designed to evaluate video quality enhancement models, with a focus on the ESRGAN (Enhanced Super-Resolution Generative Adversarial Network) model. The proposed platform assesses model performance based on accuracy, processing time, and storage efficiency. Through a systematic comparison of original and upscaled videos, this study provides an analysis of the ESRGAN model’s capabilities. The research is organized into four principal modules: downscaling video resolution to establish a baseline, upscaling using the ESRGAN model, evaluating the accuracy of upscaled video, and deploying the benchmarking platform locally. The results highlight the effectiveness of ESRGAN and lay the groundwork for future evaluations of additional video enhancement models. Future directions include expanding the platform to benchmark multiple state-of-the-art models, thereby enhancing its utility in the video processing field.