src/llama-kv-cache.cpp at main in okami.mom/llama-cpp-bonsai

skip to content

≡ llama-cpp-bonsai

101 kB 3,015 lines

904cb4ae

bonsai: suffix cache reuse (SCR) for linear attention hybrids (scr)

nekomimi.pet Oct 13

⋯

src/llama-kv-cache.cpp

at main C++