From fdf52b10b816ccdcfb799b3ced2ec65442e2ffa7 Mon Sep 17 00:00:00 2001
From: Wangshanshan <30051912+dominicshanshan@users.noreply.github.com>
Date: Mon, 17 Nov 2025 23:02:26 -0800
Subject: [PATCH] Extend openAI server waiting time to avoid large model weight
 loading out of time.

Signed-off-by: Wangshanshan <30051912+dominicshanshan@users.noreply.github.com>
---
 tests/unittest/llmapi/apps/openai_server.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tests/unittest/llmapi/apps/openai_server.py b/tests/unittest/llmapi/apps/openai_server.py
index ca98f7e1ece..9a9f38193cd 100644
--- a/tests/unittest/llmapi/apps/openai_server.py
+++ b/tests/unittest/llmapi/apps/openai_server.py
@@ -16,7 +16,7 @@
 
 class RemoteOpenAIServer:
     DUMMY_API_KEY = "tensorrt_llm"
-    MAX_SERVER_START_WAIT_S = 600  # wait for server to start for 600 seconds
+    MAX_SERVER_START_WAIT_S = 7200  # wait for server to start for 7200 seconds (~ 2 hours) for LLM models weight loading
 
     def __init__(self,
                  model: str,