Primary sequence databases (NCBI/ENA/DDBJ) have been around for decades and are always current (they share sequences overnight). They are primary repository of sequences and sync submissions overnight. They carry annotations for parts of their sequences for others there may be none.
Genome Reference Consortium is the apex body that manages genome releases for important genomes (human, mouse, zebrafish, rat and chicken). They release primary genome builds that then get deposited into appropriate sections of primary databases. Organizations offer annotation that they internally generate (NCBI, Ensembl, UCSC) but the underlying sequence is identical for a given genome build.
There are plenty of other derived/special focus databases. UniProt (LINK) is all things proteins, PDB (LINK) for protein structures.
You will find organism specific databases that originally provided sequence/annotations for those genomes around. They were useful in early days of genome sequencing but as NGS tool off they became subject to disappearing grant money. Some have turned to a subscription model (e.g. TAIR, BioCyc, KEGG) to support themselves. Parts of them may still be freely accessible but other parts require a subscription. If you are lucky enough to have access then you are all set otherwise you end up finding other (perhaps less desirable) alternatives. But you should be able to find the info you need in some free form elsewhere.