]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/commit
overview / pkg / ggml / sources / llama.cpp / commit
summary | shortlog | log | commit | commitdiff | tree
(parent: df984e0)
llama : minor fixes for up llama load model speed (#11448)
authorlexasub <redacted>
Mon, 27 Jan 2025 13:42:09 +0000 (17:42 +0400)
committerGitHub <redacted>
Mon, 27 Jan 2025 13:42:09 +0000 (14:42 +0100)
commita5203b4465c5c87813936bde98170e25bb09024f
treec5eb4d3912a1dea938bcc1a462a7083a3be68a37tree
parentdf984e014714cba4c99ef894b20b51cbcef31b16commit | diff
llama : minor fixes for up llama load model speed (#11448)

* impl::load change map bpe_ranks to onordered map for reduce time of impl::load on 30%

* llama_model_loader::init_mapping - replace new llama_mmap to std::make_unique<llama_mmap> for clean code & reduce (/2) time of running init_mappings

* Update src/llama-vocab.cpp

---------

Co-authored-by: lexasub <redacted>
Co-authored-by: Diego Devesa <redacted>
src/llama-model-loader.cpp diff | blob | history
src/llama-vocab.cpp diff | blob | history
Packaging of ggml-org/llama.cpp