Update the model and print generating response time

Dockerfile

@@ -96,8 +96,8 @@ while ! curl -s http://localhost:11434 >/dev/null; do\n\
     sleep 2\n\
 done\n\
 \n\
-echo "Pulling mixtral model..."\n\
-ollama pull mixtral\n\
+echo "Pulling gemma3 model..."\n\
+ollama pull gemma3\n\
 \n\
 echo "Starting FastAPI application..."\n\
 exec uvicorn main:app --host 0.0.0.0 --port 7860' > start.sh && \

main.py

@@ -44,12 +44,12 @@ class AIAgent:
         self.endpoints_documentation = endpoints_documentation
         self.ollama_base_url = "http://localhost:11434"  # Default Ollama URL
         # self.model_name = "mistral"  # Using mistral model for better multilingual support
-        self.model_name = 'mixtral'
-        self.BASE_URL = 'https://0a84-105-196-69-205.ngrok-free.app'
+        self.model_name = 'gemma3'
+        self.BASE_URL = 'https://a961-105-196-69-205.ngrok-free.app'
         self.headers = {
             'Content-type': 'application/json'
         }
-        self.user_id = '03564fa7-0827-43d5-ba1d-cff2177e2d09'
+        self.user_id = '76ceed74-143a-45c1-843e-ba583c122dea'
         self.max_retries = 3
         self.retry_delay = 2  # seconds
         
@@ -530,6 +530,9 @@ class AIAgent:
                 "extracted_keywords": ", ".join(extracted_keywords),
                 "sentiment_analysis": json.dumps(sentiment_result)
             })
+
+            # TODO: remove the print statement in production
+            print('End time of generating response: ', time.time() - start_time)
             
             # 2. Parse the router response
             route_result = router_result["route_result"]