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 17, 08:00 UTC
[Fix] Fix default port number in benchmark scripts (#265)
Zhuohan Li committed
Jun 26, 2023 at 13:15 UTC
43710e8d0965d616db51639ced76605dee1bde93
2 files changed
+2
-2
benchmarks/benchmark_serving.py
+1
-1
@@ -217,7 +217,7 @@ if __name__ == "__main__":
217
parser.add_argument("--backend", type=str, default="vllm",
218
choices=["vllm", "tgi"])
219
parser.add_argument("--host", type=str, default="localhost")
220
- parser.add_argument("--port", type=int, default=8001)
220
+ parser.add_argument("--port", type=int, default=8000)
221
parser.add_argument("--dataset", type=str, required=True,
222
help="Path to the dataset.")
223
parser.add_argument("--tokenizer", type=str, required=True,
benchmarks/launch_tgi_server.sh
+1
-1
@@ -1,6 +1,6 @@
1
#!/bin/bash
2
3
-PORT=8001
3
+PORT=8000
4
MODEL=$1
5
TOKENS=$2
6