You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
ggml-cuda : add rope f16, restore performance with parallel decoding#3272
Merged
ggerganov merged 4 commits intocustom-attention-maskggml-org/llama.cpp:custom-attention-maskfrom cam-cuda-2ggml-org/llama.cpp:cam-cuda-2Copy head branch name to clipboardSep 20, 2023
