]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/commit
overview / pkg / ggml / sources / llama.cpp / commit
summary | shortlog | log | commit | commitdiff | tree
(parent: b2d80e1)
llava : MobileVLM support (#4954)
authorXiaotaoChen <redacted>
Mon, 22 Jan 2024 13:09:35 +0000 (21:09 +0800)
committerGitHub <redacted>
Mon, 22 Jan 2024 13:09:35 +0000 (15:09 +0200)
commit3ce7e8f8e7ccfce07e5947ac5f1f3f4628cf68ea
tree75c5f7d2eb2e6df853fe1fa8cb7119f3178a59a3tree
parentb2d80e105a59b54822edf7ce7f3ed5f317e96e21commit | diff
llava : MobileVLM support (#4954)

* MobileVLM native implementation

* delete depthwise_conv_2d and permute_cpy relative code, replace the two by the existed functions, and opt ldp definition, support LLAMA_PERF option for CMake

* move android script to example/llava directory

* Fix the editor config checks

---------

Co-authored-by: Chenxiaotao03 <redacted>
CMakeLists.txt diff | blob | history
examples/llava/MobileVLM-README.md [new file with mode: 0644] blob
examples/llava/android/adb_run.sh [new file with mode: 0755] blob
examples/llava/android/build_64.sh [new file with mode: 0755] blob
examples/llava/clip.cpp diff | blob | history
examples/llava/convert-image-encoder-to-gguf.py diff | blob | history
ggml.c diff | blob | history
ggml.h diff | blob | history
Packaging of ggml-org/llama.cpp