specific method · filed under software implementation
Flash-Sparse-Attention
A named sparse-attention method distinct from flash-moba.
- source
- 1
- model
- 1
- labs adopt it
- 0
- strongest
- evaluated
How sources treat it
One count per evidence span, weakest treatment to strongest.
evaluated 1
Documented in
Evidence
1 span quoted from the sources, strongest treatment first.
more than 4× faster than the open-source Flash-Sparse-Attention and flash-moba
evaluatedevaluation onlyin MiniMax-M3MiniMax
Filed alongside
Other methods under software implementation :: kernel & quantization library.