A high-throughput and memory-efficient inference and serving engine for LLMs
Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git
Synced Nov 15, 03:30 UTC
[Misc] Change warn_for_unimplemented_methods to debug (#20455)
Michael Goin committed
Jul 4, 2025 at 11:35 UTC
7e1665b0899778ceeac0e4ab81e157709cf0b57e
1 file changed
+2
-2
vllm/utils/__init__.py
+2
-2
@@ -2799,7 +2799,7 @@ def warn_for_unimplemented_methods(cls: type[T]) -> type[T]:
2799
if unimplemented_methods:
2800
method_names = ','.join(unimplemented_methods)
2801
msg = (f"Methods {method_names} not implemented in {self}")
2802
- logger.warning(msg)
2802
+ logger.debug(msg)
2803
2804
@wraps(original_init)
2805
def wrapped_init(self, *args, **kwargs) -> None:
@@ -3005,4 +3005,4 @@ def has_deep_ep() -> bool:
3005
def has_deep_gemm() -> bool:
3006
"""Whether the optional `deep_gemm` package is available."""
3007
3008
- return _has_module("deep_gemm")
\ No newline at end of file
3008
+ return _has_module("deep_gemm")