cturan/llama.cpp @ 8781013ef654270cbead3e0011e33a6d690fb168

-ын хуулбар https://github.com/cturan/llama.cpp

Georgi Gerganov 8c00b7a6ff sync : ggml (Metal F32 support + reduce ggml-alloc size) (#3192)		2 жил өмнө
..
CMakeLists.txt	097e121e2f llama : add benchmark example (#2626)	2 жил өмнө
llama-bench.cpp	8c00b7a6ff sync : ggml (Metal F32 support + reduce ggml-alloc size) (#3192)	2 жил өмнө