Open Models Almanac
Methodsmodel architecturetoken mixersparse attentionblock selection

single mention · model architecture

Mixture of Block Attention

A sparse-attention method that uses a mixture of blocks as its organizing mechanism; the evidence provides no further selection details.

The sources on this page only mention it in passing. The line above is the extraction's one-line gloss. See block selection for the methods filed beside it.

Also called MoBA.

Filed under model architecture › token mixer › sparse attention › block selection