]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/commit
overview / pkg / ggml / sources / llama.cpp / commit
summary | shortlog | log | commit | commitdiff | tree
(parent: c8771ab)
Clarify default MMQ for CUDA and LLAMA_CUDA_FORCE_MMQ flag (#8115)
authorIsaac McFadyen <redacted>
Wed, 26 Jun 2024 06:29:28 +0000 (02:29 -0400)
committerGitHub <redacted>
Wed, 26 Jun 2024 06:29:28 +0000 (08:29 +0200)
commit88540445615e77a0177fcca43aaa8e9d8eea6864
tree6d5272baa14269ecc5466452a1c9ea2ad3160872tree
parentc8771ab5f89387cdd7d9a8a69280dac46b45e02fcommit | diff
Clarify default MMQ for CUDA and LLAMA_CUDA_FORCE_MMQ flag (#8115)

* Add message about int8 support

* Add suggestions from review

Co-authored-by: Johannes Gäßler <redacted>
---------

Co-authored-by: Johannes Gäßler <redacted>
README.md diff | blob | history
Packaging of ggml-org/llama.cpp