Update app.py

app.py

@@ -71,19 +71,18 @@ def generate(prompt, history=[], temperature=0.1, max_new_tokens=10000, top_p=0.
     available_tokens = 32768 - total_tokens_used
 
     if available_tokens <= 0:
-        return f"Error: 입력이 최대 허용 토큰 수를 초과합니다.\nTotal tokens used: {total_tokens_used}"
+        yield f"Error: 입력이 최대 허용 토큰 수를 초과합니다. Total tokens used: {total_tokens_used}"
+        return
 
     formatted_prompt = format_prompt(prompt, history)
     try:
         stream = client.text_generation(formatted_prompt, temperature=temperature, max_new_tokens=min(max_new_tokens, available_tokens),
                                         top_p=top_p, repetition_penalty=repetition_penalty, do_sample=True, seed=42, stream=True)
-        output = ""
         for response in stream:
-            output += response['generated_text'] if 'generated_text' in response else str(response)
-        return f"{output}\n\n---\nTotal tokens used: {total_tokens_used}"
+            output = response['generated_text'] if 'generated_text' in response else str(response)
+            yield f"{output}\n\n---\nTotal tokens used: {total_tokens_used}"
     except Exception as e:
-        return f"Error: {str(e)}\nTotal tokens used: {total_tokens_used}"
-        
+        yield f"Error: {str(e)}\nTotal tokens used: {total_tokens_used}"
 
 mychatbot = gr.Chatbot(
     avatar_images=["./user.png", "./botm.png"],