Skip to content

v1.10.1

Choose a tag to compare

@oobabooga oobabooga released this 13 Jul 17:56
0315122

Library updates

  • FlashAttention: bump to v2.6.1. Now Gemma-2 works in ExLlamaV2 with FlashAttention without any quality loss.

Bug fixes

  • Fix for MacOS users encountering model load errors with llama.cpp (#6227). Thanks @InvectorGator.