Update app.py

app.py

@@ -74,7 +74,7 @@ def chat_completions():
         full_response = ""
         while True:
             conversation = "\n".join([f"{msg['role']}: {msg['content']}" for msg in messages])
-            conversation += "\n请关注并回复user最近的消息并避免总结对话历史的回答"
+            conversation += "\nPlease follow and reply to the user’s recent messages and avoid answers that summarize the conversation history."
             
             payload = {
                 "text": conversation,
@@ -101,19 +101,19 @@ def chat_completions():
                     
                     if finish_reason == 'length':
                         messages.append({"role": "assistant", "content": full_response})
-                        messages.append({"role": "user", "content": "请继续你的输出，不要重复之前的内容"})
-                        break  # 跳出当前循环，继续下一次请求
+                        messages.append({"role": "user", "content": "Please continue your output and do not repeat the previous content"})
+                        break  # Jump out of the current loop and continue with the next request
                     else:
-                        # 正常结束，发送最后的内容（如果有的话）
+                        # End normally, sending the final content (if any)
                         last_content = response_message.get('text', '')
                         if last_content and last_content != full_response:
                             yield f"data: {json.dumps(format_openai_response(last_content[len(full_response):]))}\n\n"
                         
                         yield f"data: {json.dumps(format_openai_response('', finish_reason))}\n\n"
                         yield "data: [DONE]\n\n"
-                        return  # 完全结束生成
+                        return  # completely end generation
 
-        # 如果因为多次长度限制而最终结束，发送一个停止信号
+        # If it ends due to multiple length limits, send a stop signal
         yield f"data: {json.dumps(format_openai_response('', 'stop'))}\n\n"
         yield "data: [DONE]\n\n"