t

llm_server/routes/openai/chat_completions.py

@@ -57,6 +57,7 @@ def openai_chat_completions():
             else:
                 handler.prompt = transform_messages_to_prompt(handler.request.json['messages'])
 
+            generated_text = ''
             response_status_code = 0
             start_time = time.time()
 
@@ -154,3 +155,4 @@ def openai_chat_completions():
                     # The worker incremented it, we'll decrement it.
                     decrement_ip_count(handler.client_ip, 'processing_ips')
                     decr_active_workers(handler.selected_model, handler.backend_url)
+                    print(len(generated_text))