Update app.py

app.py

@@ -30,7 +30,12 @@ def get_messages_formatter_type(model_name):
 def chat_fn(message, history, model, system_message, max_tokens, temperature, top_p, top_k, repeat_penalty):
     history_list = history or []
     response_generator = respond(message, history_list, model, system_message, max_tokens, temperature, top_p, top_k, repeat_penalty)
-    return response_generator, history_list
+    
+    full_response = ""
+    for current_history in response_generator:
+        full_response = current_history[-1][1]  # 获取最新的回复
+        yield current_history, history #  在每次迭代中返回当前历史记录和历史记录
+
 
 def respond(
     message,
@@ -136,7 +141,7 @@ with gr.Blocks(theme=gr.themes.Soft(primary_hue="violet", secondary_hue="violet"
         code_background_fill_dark="#292733",
     )) as demo:
 
-    chatbot = gr.Chatbot(scale=1, show_copy_button=True)
+    chatbot = gr.Chatbot(scale=1, show_copy_button=True, type='messages') #  修改 chatbot 类型
     message = gr.Textbox(label="Your message")
     model_dropdown = gr.Dropdown(
         ["openbuddy-llama3.2-3b-v23.2-131k-q5_k_m-imat.gguf"],
@@ -150,12 +155,8 @@ with gr.Blocks(theme=gr.themes.Soft(primary_hue="violet", secondary_hue="violet"
     top_p = gr.Slider(minimum=0.1, maximum=2.0, value=0.9, step=0.05, label="Top-p")
     top_k = gr.Slider(minimum=0, maximum=100, value=1, step=1, label="Top-k")
     repeat_penalty = gr.Slider(minimum=0.0, maximum=2.0, value=1.1, step=0.1, label="Repetition penalty")
-
     history = gr.State([])
 
-    def chat_fn(message, history, model, system_message, max_tokens, temperature, top_p, top_k, repeat_penalty):
-        return respond(message, history, model, system_message, max_tokens, temperature, top_p, top_k, repeat_penalty)
-
     message.submit(chat_fn, [message, history, model_dropdown, system_message, max_tokens, temperature, top_p, top_k, repeat_penalty], [chatbot, history])
 
     gr.Markdown(description)