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 14, 05:30 UTC
[Refactor] Remove Unused Environment Variable `VLLM_NO_DEPRECATION_WARNING` (#22199)
Signed-off-by: yewentao256 <zhyanwentao@126.com>
Wentao Ye committed
Aug 5, 2025 at 12:40 UTC
ae87ddd040b793fd9f4f05cb660a4728c81d7670
1 file changed
-5
vllm/envs.py
-5
@@ -70,7 +70,6 @@ if TYPE_CHECKING:
70
NVCC_THREADS: Optional[str] = None
71
VLLM_USE_PRECOMPILED: bool = False
72
VLLM_TEST_USE_PRECOMPILED_NIGHTLY_WHEEL: bool = False
73
- VLLM_NO_DEPRECATION_WARNING: bool = False
73
VLLM_KEEP_ALIVE_ON_ENGINE_DEATH: bool = False
74
CMAKE_BUILD_TYPE: Optional[str] = None
75
VERBOSE: bool = False
@@ -582,10 +581,6 @@ environment_variables: dict[str, Callable[[], Any]] = {
581
lambda: bool(
582
int(os.getenv("VLLM_ENABLE_FUSED_MOE_ACTIVATION_CHUNKING", "1"))),
583
585
- # If set, vllm will skip the deprecation warnings.
586
- "VLLM_NO_DEPRECATION_WARNING":
587
- lambda: bool(int(os.getenv("VLLM_NO_DEPRECATION_WARNING", "0"))),
588
-
584
# If set, the OpenAI API server will stay alive even after the underlying
585
# AsyncLLMEngine errors and stops serving requests
586
"VLLM_KEEP_ALIVE_ON_ENGINE_DEATH":