llama ver 3 model breaking changes, 7% speedup though

#4
by Tom-Neverwinter - opened

This is not a GGML repository.

TehVenom changed discussion status to closed

Sign up or log in to comment