Update server.py

server.py

@@ -25,13 +25,13 @@ def launch_vllm_server(beta=1.0):
         "--max-num-seqs", "1",
         "--port", "8000"
     ]
-    subprocess.run(cmd, env=env)
+    _server_process = subprocess.Popen(cmd, env=env)
+        
+    # Wait for the process to complete or be terminated
+    _server_process.wait()
 
 # Step 1: Setup
 setup_mixinputs()
 
 # Step 2: Launch vLLM server in background
 threading.Thread(target=launch_vllm_server, daemon=True).start()
-
-# Step 3: Give time for server to initialize
-time.sleep(60)