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, 19:45 UTC
Fix typo in benchmark_serving_structured_output.py (#14566)
Signed-off-by: Russell Bryant <rbryant@redhat.com>
Russell Bryant committed
Mar 10, 2025 at 17:58 UTC
432d6dad15e74ba2d5f2f97f9693706b0930b2f0
1 file changed
+5
-6
benchmarks/benchmark_serving_structured_output.py
+5
-6
@@ -989,12 +989,11 @@ if __name__ == "__main__":
989
type=float,
990
default=1.0,
991
help="Ratio of Structured Outputs requests")
992
- parser.add_argument(
993
- "--structured-output-backend",
994
- type=str,
995
- choices=["outlines", "lm-format-enforcer", "xgrammar", "json-unique"],
996
- default="xgrammar",
997
- help="Backend to use for structured outputs")
992
+ parser.add_argument("--structured-output-backend",
993
+ type=str,
994
+ choices=["outlines", "lm-format-enforcer", "xgrammar"],
995
+ default="xgrammar",
996
+ help="Backend to use for structured outputs")
997
998
args = parser.parse_args()
999
main(args)