Open Models Almanac
Methodsmodel architecturemultimodal architecture

single mention · model architecture

Single-matmul vision projection

Replaces a vision encoder with a single large matrix multiplication for visual projection.

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

Also called single large matmul projection for vision.

Filed under model architecture › multimodal architecture