Open Models Almanac
Methodsinference & servinginference kernel

single mention · inference & serving

FlexAttention

FlexAttention is used to optimize computation for sparse multimodal fusion.

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

Also called FlexAttention for sparse multimodal fusion.

Filed under inference & serving › inference kernel