I tried the 72b v2 Q6 gguf model and it seems to underperform a lot in my task compared to the original model gguf.
· Sign up or log in to comment