Model techniques map
Techniques

implementation detail · not yet filed

Attention memory-cost optimization in the Triton implementation

source
1
model
1
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.

It also has some optimization on the attention code to reduce the memory cost.

usedsoftware implementationin gpt-oss reference implementationsOpenAI