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 16, 04:30 UTC
[Misc]Suppress irrelevant exception stack trace information when CUDA… (#11438)
Co-authored-by: shiquan <shiquan>
dpxa committed
Dec 24, 2024 at 16:43 UTC
4f074fbf53f9e11a57a4c3d8b084796d155a270a
1 file changed
+2
-1
vllm/entrypoints/llm.py
+2
-1
@@ -233,7 +233,8 @@ class LLM:
233
self.request_counter = Counter()
234
235
def __del__(self):
236
- if self.llm_engine and hasattr(self.llm_engine, "shutdown"):
236
+ if hasattr(self, 'llm_engine') and self.llm_engine and hasattr(
237
+ self.llm_engine, "shutdown"):
238
self.llm_engine.shutdown()
239
240
@staticmethod