]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/commit
overview / pkg / ggml / sources / llama.cpp / commit
summary | shortlog | log | commit | commitdiff | tree
(parent: 8f429fa)
speculative : PoC for speeding-up inference via speculative sampling (#2926)
authorGeorgi Gerganov <redacted>
Sun, 3 Sep 2023 12:12:08 +0000 (15:12 +0300)
committerGitHub <redacted>
Sun, 3 Sep 2023 12:12:08 +0000 (15:12 +0300)
commit47068e517004d90f13c16352bb3b4cafd53a00cd
tree259f1fb1184775dc250452d319c8006c0704ea22tree
parent8f429fa5111901f9646cf998643ac5310846d487commit | diff
speculative : PoC for speeding-up inference via speculative sampling (#2926)

* speculative : initial example

* speculative : print encoding speed

* speculative : add --draft CLI arg
common/common.cpp diff | blob | history
common/common.h diff | blob | history
examples/CMakeLists.txt diff | blob | history
examples/main/main.cpp diff | blob | history
examples/speculative/CMakeLists.txt [new file with mode: 0644] blob
examples/speculative/speculative.cpp [new file with mode: 0644] blob
Packaging of ggml-org/llama.cpp