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 12, 09:45 UTC
Revert "Supress verbose logs from model_hosting_container_standards (… (#29335)
Signed-off-by: Shen Teng <sheteng@amazon.com> Co-authored-by: Cyrus Leung <tlleungac@connect.ust.hk>
HappyAmazonian committed
Nov 28, 2025 at 05:29 UTC
f8151b66fa23f79c470d639f34768eb37709df0a
2 files changed
+1
-5
requirements/common.txt
+1
-1
@@ -49,4 +49,4 @@ cbor2 # Required for cross-language serialization of hashable objects
49
setproctitle # Used to set process names for better debugging and monitoring
50
openai-harmony >= 0.0.3 # Required for gpt-oss
51
anthropic == 0.71.0
52
-model-hosting-container-standards < 1.0.0
\ No newline at end of file
52
+model-hosting-container-standards >= 0.1.9, < 1.0.0
\ No newline at end of file
vllm/entrypoints/openai/api_server.py
-4
@@ -5,7 +5,6 @@ import hashlib
5
import importlib
6
import inspect
7
import json
8
-import logging
8
import multiprocessing
9
import multiprocessing.forkserver as forkserver
10
import os
@@ -2099,9 +2098,6 @@ async def run_server(args, **uvicorn_kwargs) -> None:
2098
# Add process-specific prefix to stdout and stderr.
2099
decorate_logs("APIServer")
2100
2102
- # Suppress verbose logs from model_hosting_container_standards
2103
- logging.getLogger("model_hosting_container_standards").setLevel(logging.ERROR)
2104
-
2101
listen_address, sock = setup_server(args)
2102
await run_server_worker(listen_address, sock, args, **uvicorn_kwargs)
2103