The I-TASSER server is an online resource for automated protein structure prediction and structure-based function annotation. It uses multiple threading alignment approaches to recognize structural templates from the PDB, constructs full-length structure models through iterative fragment assembly simulations, and derives functional insights by matching predicted structure models with known proteins in function databases. Recent developments include new methods for atomic-level structure refinement, local structure quality estimation, and biological function annotations. These improvements aim to enhance the accuracy and usability of the server. The server has been widely used for various biological and biomedical investigations, and user feedback has led to significant updates. The I-TASSER pipeline involves three steps: template identification, full-length structure assembly, and structure-based function annotation. The server has generated structure models and functional predictions for over 200,000 proteins from more than 50,000 users in 117 countries. Recent developments include a new approach to estimate residue-level local quality, B-factor prediction, atomic-level structure refinement, a consensus-based ligand-binding site prediction, integration of the BioLiP function library, and a new message board system for user communication. The server's performance has been validated in benchmark and blind tests, showing high accuracy in protein structure and function prediction. The I-TASSER server is a cutting-edge tool for automated protein structure and function prediction, continuously improved based on user feedback and method development. The server's template libraries are updated weekly and freely available. Users are encouraged to report issues and discuss the server on the message board. The I-TASSER server is designed to provide accurate and useful structure and function predictions based on state-of-the-art methods.The I-TASSER server is an online resource for automated protein structure prediction and structure-based function annotation. It uses multiple threading alignment approaches to recognize structural templates from the PDB, constructs full-length structure models through iterative fragment assembly simulations, and derives functional insights by matching predicted structure models with known proteins in function databases. Recent developments include new methods for atomic-level structure refinement, local structure quality estimation, and biological function annotations. These improvements aim to enhance the accuracy and usability of the server. The server has been widely used for various biological and biomedical investigations, and user feedback has led to significant updates. The I-TASSER pipeline involves three steps: template identification, full-length structure assembly, and structure-based function annotation. The server has generated structure models and functional predictions for over 200,000 proteins from more than 50,000 users in 117 countries. Recent developments include a new approach to estimate residue-level local quality, B-factor prediction, atomic-level structure refinement, a consensus-based ligand-binding site prediction, integration of the BioLiP function library, and a new message board system for user communication. The server's performance has been validated in benchmark and blind tests, showing high accuracy in protein structure and function prediction. The I-TASSER server is a cutting-edge tool for automated protein structure and function prediction, continuously improved based on user feedback and method development. The server's template libraries are updated weekly and freely available. Users are encouraged to report issues and discuss the server on the message board. The I-TASSER server is designed to provide accurate and useful structure and function predictions based on state-of-the-art methods.