Update app.py

app.py

@@ -89,6 +89,7 @@ with gr.Blocks(css="style.css") as demo:
     def generate(message, history, max_new_tokens: int, temperature: float):        
         generation_config = {"max_new_tokens": max_new_tokens, "temperature": temperature}
         inference = pipe(sequences=message, streaming=True, **generation_config)
+        history[-1][1] += message
         for token in inference:
             history[-1][1] += token.generations[0].text
             yield history