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 10, 07:45 UTC
[Misc] Disable default `--ready-check-timeout-sec` extra call in vllm bench (#30975)
Signed-off-by: NickLucche <nlucches@redhat.com>
Nicolò Lucchesi committed
Jan 12, 2026 at 10:58 UTC
22970c16266746971f80fbd8e1fd505a5f35511d
1 file changed
+2
-3
vllm/benchmarks/serve.py
+2
-3
@@ -1488,10 +1488,9 @@ def add_cli_args(parser: argparse.ArgumentParser):
1488
parser.add_argument(
1489
"--ready-check-timeout-sec",
1490
type=int,
1491
- default=600,
1491
+ default=0,
1492
help="Maximum time to wait for the endpoint to become ready "
1493
- "in seconds (default: 600 seconds / 10 minutes). If set to 0, "
1494
- "the ready check will be skipped.",
1493
+ "in seconds. Ready check will be skipped by default.",
1494
)
1495
1496
parser.add_argument(