jadidbourbaki/llama.cpp
The drafting loop in common/ngram-cache.cpp copied an inner common_ngram_cache_part map in three places on every drafting step. I changed all three to read the part through a const reference.
try_...
jadidbourbaki/llama.cpp PR #2:
LLM inference in C/C++. Contribute to jadidbourbaki/llama.cpp development by creating an account on GitHub.
Comments (0)
No comments yet