Joosua Santasalo

icon indicating a website link securecloud.blog

icon location Finland icon location Former Microsoft MVP 20-25 (july), Former MSRC top100 most valuable researcher 23-24 - fooling around with NodeJS and Azure - Working with Secureworks (Sophos)

Results 13 comments of Joosua Santasalo

Optimization: change userMapper to use Memberof instead of enumeration group members

Using all group members is still beneficial for certain future features (such as expanding first 10 group members for inspection)

Massive performance regression on 0.1.32 -> GGML_CUDA_FORCE_MMQ: (SET TO NO, after 0.1.31)

Thx for the help. I tried today on A6000 and the issue is same for embedding, so I believe it is specific to the task but impacts most NVIDIA architectures....

Massive performance regression on 0.1.32 -> GGML_CUDA_FORCE_MMQ: (SET TO NO, after 0.1.31)

I was able to sucessfully build with the flags, but see only minimal performance improvement :) will need to see for any other differences, what changes after 31 ->