[server] Passing model_override_args to launch_server via the CLI. (#1298)

Signed-off-by: Kai-Hsun Chen <kaihsun@anyscale.com>
2024-09-09 02:14:25 -07:00
parent 662ecd9368
commit c9b75917d5
8 changed files with 71 additions and 16 deletions
--- a/test/srt/test_serving_latency.py
+++ b/test/srt/test_serving_latency.py
@@ -12,7 +12,7 @@ class TestServingLatency(unittest.TestCase):
            "python3",
            "-m",
            "sglang.bench_latency",
-            "--model",
+            "--model-path",
            DEFAULT_MODEL_NAME_FOR_TEST,
            "--batch-size",
            "1",