]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/commit
overview / pkg / ggml / sources / llama.cpp / commit
summary | shortlog | log | commit | commitdiff | tree
(parent: dba497e)
llava : add support for moondream vision language model (#6899)
authorvik <redacted>
Thu, 25 Apr 2024 19:38:31 +0000 (12:38 -0700)
committerGitHub <redacted>
Thu, 25 Apr 2024 19:38:31 +0000 (22:38 +0300)
commit46e12c4692a37bdd31a0432fc5153d7d22bc7f72
tree39c8399c7df13bd49493e6b003aad1568621f367tree
parentdba497e0c1eff27cf0e85d1d73c86fa08e1b5557commit | diff
llava : add support for moondream vision language model (#6899)

* add support for moondream vision language model

This required making the following changes to the CLIP model:

1. Support for patch embedding bias.
2. Make class embedding and pre-layernorm optional.
3. Add support for post-layernorm.

* Update examples/llava/clip.cpp

---------

Co-authored-by: Georgi Gerganov <redacted>
README.md diff | blob | history
examples/llava/clip.cpp diff | blob | history
Packaging of ggml-org/llama.cpp