@VivekLeon can you check this? It works for me (tested the q8q4 version with latest llama.cpp)
· Sign up or log in to comment
