The Pfam database has undergone significant reorganization to reduce the effort required for each release, enabling more frequent updates. The most notable change is the shift from using the full UniProtKB database to reference proteomes, which provides a more stable and manageable set of sequences. This change reduces the need for manual curation and simplifies the sequence database size. Pfam now primarily uses reference proteomes, with some entries still matching the full UniProtKB database. The removal of Pfam-B, an automatically generated supplement, has also been implemented. The current release (Pfam 29.0) includes 16,295 entries and 559 clans, with improvements in the visualization of family relationships within clans. The transition to reference proteomes and a more relaxed overlap rule have streamlined the curation process, reducing the time required for each release. These changes aim to enhance the sustainability and efficiency of Pfam's production while maintaining high-quality annotations.The Pfam database has undergone significant reorganization to reduce the effort required for each release, enabling more frequent updates. The most notable change is the shift from using the full UniProtKB database to reference proteomes, which provides a more stable and manageable set of sequences. This change reduces the need for manual curation and simplifies the sequence database size. Pfam now primarily uses reference proteomes, with some entries still matching the full UniProtKB database. The removal of Pfam-B, an automatically generated supplement, has also been implemented. The current release (Pfam 29.0) includes 16,295 entries and 559 clans, with improvements in the visualization of family relationships within clans. The transition to reference proteomes and a more relaxed overlap rule have streamlined the curation process, reducing the time required for each release. These changes aim to enhance the sustainability and efficiency of Pfam's production while maintaining high-quality annotations.