Model techniques map
Techniquesinference & servinginference kernel

specific method · filed under inference & serving

DeepGEMM-based batch-invariant matrix multiplication

Replacing the matrix-multiplication implementation end-to-end with DeepGEMM in the batch-invariant kernel context.

Also called replace it end-to-end with DeepGEMM.

source
1
models
2
lab adopt it
1
strongest
used

How sources treat it

One count per evidence span, weakest treatment to strongest.

used 1

Documented in

Evidence

1 span quoted from the sources, strongest treatment first.

Therefore, we replace it end-to-end with DeepGEMM.

usedunclearin DeepSeek-V4DeepSeek

Filed alongside

Other methods under inference & serving :: inference kernel.