diff --git a/llama_cpp/server/app.py b/llama_cpp/server/app.py index 3e45684..f46f920 100644 --- a/llama_cpp/server/app.py +++ b/llama_cpp/server/app.py @@ -48,6 +48,9 @@ class Settings(BaseSettings): vocab_only: bool = Field( default=False, description="Whether to only return the vocabulary." ) + verbose: bool = Field( + default=True, description="Whether to print debug information." + ) router = APIRouter() @@ -83,6 +86,7 @@ def create_app(settings: Optional[Settings] = None): n_ctx=settings.n_ctx, last_n_tokens_size=settings.last_n_tokens_size, vocab_only=settings.vocab_only, + verbose=settings.verbose, ) if settings.cache: cache = llama_cpp.LlamaCache()