Update app.py

app.py

@@ -388,7 +388,7 @@ class LlamaGenerator(BaseGenerator):
         )
         
         # Initialize models
-        self.model_manager.load_model(
+        self.model_manager.load_model( 
             "llama",
             llama_model_name,
             "llama",
@@ -448,6 +448,52 @@ class LlamaGenerator(BaseGenerator):
             model_kwargs,
             **kwargs
         )
+
+
+    def generate_with_context(
+        self,
+        context: str,
+        user_input: str,
+        chat_history: List[Tuple[str, str]],
+        model_kwargs: Dict[str, Any],
+        max_history_turns: int = 3,
+        strategy: str = "default",
+        num_samples: int = 5,
+        depth: int = 3,
+        breadth: int = 2,
+        
+    ) -> str:
+        """Generate a response using context and chat history.
+        
+        Args:
+            context (str): Context for the conversation
+            user_input (str): Current user input
+            chat_history (List[Tuple[str, str]]): List of (user, assistant) message pairs
+            model_kwargs (dict): Additional arguments for model.generate()
+            max_history_turns (int): Maximum number of history turns to include
+            strategy (str): Generation strategy
+            num_samples (int): Number of samples for applicable strategies
+            depth (int): Depth for DVTS strategy
+            breadth (int): Breadth for DVTS strategy
+            
+        Returns:
+            str: Generated response
+        """
+        prompt = self._construct_prompt(
+            context,
+            user_input,
+            chat_history,
+            max_history_turns
+        )
+        return self.generate(
+            prompt,
+            model_kwargs,
+            strategy,
+            num_samples,
+            depth,
+            breadth
+        )
+
     
     def check_health(self) -> HealthStatus:
         """Check the health status of the generator."""