Update app.py

app.py

@@ -87,7 +87,9 @@ def start_vllm_service() -> bool:
             [
                 "vllm", "serve", VLLM_MODEL_NAME,
                 "--port", str(VLLM_PORT),
-                "--host", VLLM_HOST
+                "--host", VLLM_HOST,
+                "--tensor-parallel-size", "1",
+                "--gpu-memory-utilization", "0.95"
             ],
             text=True
         )
@@ -485,7 +487,7 @@ def parse_arguments():
     parser.add_argument("--openai_api_source", default="we_provided", choices=["we_provided", "user_provided"], help="Source of OpenAI API key.")
     return parser.parse_args()
 
-
+@spaces.GPU(duration=300)
 def solve_problem_gradio(user_query, max_steps=10, max_time=60, llm_model_engine=None, enabled_tools=None):
     """
     Wrapper function to connect the solver to Gradio.