Hugging Face has announced a new feature called 'Evaluation on the Hub', which allows users to evaluate and compare models directly within their model hub. This feature is designed to simplify the process of testing and validating models, making it easier for developers to find and use high-quality models. The Evaluation on the Hub feature is based on popular evaluation libraries such as Hugging Face's Transformers library and the PyTorch framework. --- Why it matters: This matters because it simplifies the process of model evaluation and comparison, allowing researchers and engineers to quickly identify and utilize top-performing models for their specific tasks. Source: https://huggingface.co/blog/eval-on-the-hub