The SWISS-MODEL Repository (SMR) is a database of annotated 3D protein structure models generated by the automated SWISS-MODEL homology modeling pipeline. It currently holds over 400,000 high-quality models covering nearly 20% of Swiss-Prot/UniProtKB entries. This paper presents recent features and functionalities of SMR, including improved target coverage, model quality estimates, functional annotations, and enhanced in-page visualization. A new update concept has been introduced, allowing regular updates of an expanded set of core organism models and UniProtKB-based targets, complemented by user-driven on-demand updates of individual models. The new release of the modeling pipeline includes a REST API and an open licensing model for accessing model coordinates, enabling bulk downloads and re-use of models in other contexts. SMR can be accessed at https://swissmodel.expasy.org/repository. The SWISS-MODEL Repository aims to provide access to an up-to-date collection of high-quality annotated 3D protein models and experimental structure information for relevant model organism proteomes and other sequences from UniProtKB. Homology models are generated using the automated SWISS-MODEL homology modeling server pipeline, and experimental structures of proteins are mapped between the PDB and UniProtKB using SIFTS. The repository is continuously updated, with proteins from model organism proteomes updated regularly and other UniProtKB entries updated interactively by users. New models requested by users are generated by the automated SWISS-MODEL modeling pipeline and released every 15 minutes. The pipeline prioritizes templates that maximize model quality and target coverage. When templates mapping to the same sequence segment exhibit significant conformational differences, multiple models are generated to reflect structural diversity. The SWISS-MODEL Template Library (SMTL) provides a curated set of atomic coordinates from experimental structures in the PDB. It includes annotations for ligands, sequence profiles, secondary structure, solvent accessibility, and more. SMTL is searchable by BLAST and HHBlits and allows structural clustering and superposition of individual chains. SMTL entries are organized as quaternary structure assemblies and are updated weekly after each new PDB release. The modeling engine, PROMOD3, has been improved to generate model coordinates based on input alignments. It uses an optimized database of loops and the Cyclic Coordinate Descent (CCD) method for loop modeling. Sidechain modeling is inspired by the work of the Dunbrack lab and uses a backbone-dependent rotamer library. Energy minimization is performed using the OpenMM molecular mechanics library. SMR provides a continuous model database update with a three-tiered policy: continuously on-demand, weekly, and monthly. Monthly updates are initiated when a new version of UniProtKB is released, while weekly updates are initiated when the PDB releases new structures. Users can request interactive modeling jobs on the SWISS-MODEL Workspace. SMRThe SWISS-MODEL Repository (SMR) is a database of annotated 3D protein structure models generated by the automated SWISS-MODEL homology modeling pipeline. It currently holds over 400,000 high-quality models covering nearly 20% of Swiss-Prot/UniProtKB entries. This paper presents recent features and functionalities of SMR, including improved target coverage, model quality estimates, functional annotations, and enhanced in-page visualization. A new update concept has been introduced, allowing regular updates of an expanded set of core organism models and UniProtKB-based targets, complemented by user-driven on-demand updates of individual models. The new release of the modeling pipeline includes a REST API and an open licensing model for accessing model coordinates, enabling bulk downloads and re-use of models in other contexts. SMR can be accessed at https://swissmodel.expasy.org/repository. The SWISS-MODEL Repository aims to provide access to an up-to-date collection of high-quality annotated 3D protein models and experimental structure information for relevant model organism proteomes and other sequences from UniProtKB. Homology models are generated using the automated SWISS-MODEL homology modeling server pipeline, and experimental structures of proteins are mapped between the PDB and UniProtKB using SIFTS. The repository is continuously updated, with proteins from model organism proteomes updated regularly and other UniProtKB entries updated interactively by users. New models requested by users are generated by the automated SWISS-MODEL modeling pipeline and released every 15 minutes. The pipeline prioritizes templates that maximize model quality and target coverage. When templates mapping to the same sequence segment exhibit significant conformational differences, multiple models are generated to reflect structural diversity. The SWISS-MODEL Template Library (SMTL) provides a curated set of atomic coordinates from experimental structures in the PDB. It includes annotations for ligands, sequence profiles, secondary structure, solvent accessibility, and more. SMTL is searchable by BLAST and HHBlits and allows structural clustering and superposition of individual chains. SMTL entries are organized as quaternary structure assemblies and are updated weekly after each new PDB release. The modeling engine, PROMOD3, has been improved to generate model coordinates based on input alignments. It uses an optimized database of loops and the Cyclic Coordinate Descent (CCD) method for loop modeling. Sidechain modeling is inspired by the work of the Dunbrack lab and uses a backbone-dependent rotamer library. Energy minimization is performed using the OpenMM molecular mechanics library. SMR provides a continuous model database update with a three-tiered policy: continuously on-demand, weekly, and monthly. Monthly updates are initiated when a new version of UniProtKB is released, while weekly updates are initiated when the PDB releases new structures. Users can request interactive modeling jobs on the SWISS-MODEL Workspace. SMR