upd

llama_index_agent.py

@@ -17,7 +17,8 @@ class LlamaIndexAgent:
         # model_name: str = "meta-llama/llama-4-maverick:free",
         # model_name: str = "meta-llama/llama-4-scout:free",
         # model_name: str = "microsoft/phi-4-reasoning-plus:free",
-        model_name: str = "google/gemini-2.5-flash-preview",
+        # model_name: str = "google/gemini-2.5-flash-preview",
+        model_name: str = "x-ai/grok-3-beta",
         temperature: float = 0.7,
         verbose: bool = True,
     ):
@@ -132,14 +133,6 @@ class LlamaIndexAgent:
             """,
         )
 
-        self.small_agent = ReActAgent.from_tools(
-            [final_answer_tool],
-            llm=self.llm,
-            verbose=verbose,
-            max_iterations=5,
-            system_prompt="You are approached to prepare answer for the user question in desired format. You always need to use final_answer tool, it will help you.",
-        )
-
     def __call__(self, query_text: str, **kwds) -> str:
         """
         Process a user query through the agent.
@@ -154,9 +147,9 @@ class LlamaIndexAgent:
             response = self.agent.chat(query_text).response
         except:
             response = ""
-        final_response = self.small_agent.chat(f"Response: {response}")
+        final_response = tools.final_answer(answer=response)
 
-        return final_response.response
+        return final_response
 
 
 if __name__ == "__main__":

tools.py

@@ -309,7 +309,10 @@ def final_answer(answer: str, **kwargs) -> str:
     Return:
         The final answer.
     """
-    return f"""
+    resp = llm.chat(
+        messages=[
+            ChatMessage(
+                content=f"""
     Final answer from agent: {answer}
     Make final answer as short as possible.
     Final answer should be a number or as few words as possible or a comma separated list of numbers and/or strings. If you are asked for a number, don't use comma to write your number neither use units such as $ or percent sign unless specified otherwise. If you are asked for a string, don't use articles, neither abbreviations (e.g. for cities), and write the digits in plain text unless specified otherwise. If you are asked for a comma separated list, apply the above rules depending of whether the element to be put in the list is a number or a string.
@@ -317,6 +320,10 @@ def final_answer(answer: str, **kwargs) -> str:
     And on the other hand, the question might request some exact string value. Don't explain it, just return this value (For example, insted of `In response to the question, desired person is X` return only `X`)
     Again, you don't need to modify or solve answer, you just need to format it properly.
     """
+            )
+        ]
+    )
+    return resp.message.content
 
 
 # print(wikipedia_search("Mercedes Sosa studio albums"))