llama.cpp
f9ca5dcb - llama : avoid ggml_cast, use F32 query

Commit
1 year ago
llama : avoid ggml_cast, use F32 query
Author
Parents
Loading