added custom handler v8

handler.py

@@ -93,12 +93,12 @@ class EndpointHandler:
             output_text = self.tokenizer.batch_decode(output_tokens)[0]
             
             # Extract the assistant's response by removing the input prompt
-            response = output_text[len(prompt):].strip()
+            response = output_text#[len(prompt):].strip()
             logger.info(f"Generated response length: {len(response)}")
             logger.info(f"Generated response: {response}")
             
             #return [{"role": "assistant", "content": response}]
-            return response
+            return [{"result": response, "error": None}] 
             
         except Exception as e:
             logger.error(f"Error during generation: {str(e)}", exc_info=True)