Update app.py

app.py

@@ -26,6 +26,9 @@ class LLMManager:
         self.generation_lock = threading.Lock()
         self.load_all_models()
         self.session = requests.Session()
+        adapter = requests.adapters.HTTPAdapter(pool_connections=2, pool_maxsize=2)
+        self.session.mount('http://', adapter)
+        self.session.mount('https://', adapter)
 
     def load_all_models(self):
         """Cargar todos los modelos en RAM"""