diff options
| author | Kawrakow <48489457+ikawrakow@users.noreply.github.com> | 2024-07-27 07:55:01 +0200 |
|---|---|---|
| committer | GitHub <noreply@github.com> | 2024-07-27 07:55:01 +0200 |
| commit | 154e0d75fccf1784fe9ff6fd76a630b66563da3d (patch) | |
| tree | 81ce6dbb5b1900c1aa78a879f0593c694cab9d27 /docs/development/llama-star | |
| parent | 0684c3e9c70d49323b4fc517128cbe222cab7f96 (diff) | |
Merge mainline llama.cpp (#3)
* Merging mainline - WIP
* Merging mainline - WIP
AVX2 and CUDA appear to work.
CUDA performance seems slightly (~1-2%) lower as it is so often
the case with llama.cpp/ggml after some "improvements" have been made.
* Merging mainline - fix Metal
* Remove check
---------
Co-authored-by: Iwan Kawrakow <iwan.kawrakow@gmail.com>
Diffstat (limited to 'docs/development/llama-star')
| -rwxr-xr-x | docs/development/llama-star/idea-arch.key | bin | 0 -> 488591 bytes |
| -rw-r--r-- | docs/development/llama-star/idea-arch.pdf | bin | 0 -> 42334 bytes |
2 files changed, 0 insertions, 0 deletions
diff --git a/docs/development/llama-star/idea-arch.key b/docs/development/llama-star/idea-arch.key Binary files differnew file mode 100755 index 00000000..3e068e70 --- /dev/null +++ b/docs/development/llama-star/idea-arch.key diff --git a/docs/development/llama-star/idea-arch.pdf b/docs/development/llama-star/idea-arch.pdf Binary files differnew file mode 100644 index 00000000..4fa92c71 --- /dev/null +++ b/docs/development/llama-star/idea-arch.pdf |