cleanup instruct app. add queue so UI notifies users

chat.py

@@ -65,9 +65,7 @@ start_message = """
 """
 
 
-blocks = gr.Blocks()
-
-with blocks:
+with gr.Blocks() as blocks:
     gr.Markdown("# GGML Spaces Demo")
 
     chatbot = gr.Chatbot()
@@ -107,12 +105,12 @@ with blocks:
     clear.click(lambda: None, None, chatbot, queue=False)
 
     submit_click_event = submit.click(
-        fn=user, inputs=[message, chat_history_state], outputs=[message, chat_history_state], queue=False
+        fn=user, inputs=[message, chat_history_state], outputs=[message, chat_history_state], queue=True
     ).then(
         fn=chat, inputs=[chat_history_state, system_msg, max_tokens, temperature, top_p, top_k, repeat_penalty], outputs=[chatbot, chat_history_state], queue=True
     )
     message_submit_event = message.submit(
-        fn=user, inputs=[message, chat_history_state], outputs=[message, chat_history_state], queue=False
+        fn=user, inputs=[message, chat_history_state], outputs=[message, chat_history_state], queue=True
     ).then(
         fn=chat, inputs=[chat_history_state, system_msg, max_tokens, temperature, top_p, top_k, repeat_penalty], outputs=[chatbot, chat_history_state], queue=True
     )

app.py → instruct.py

@@ -23,11 +23,10 @@ description = f"""llama.cpp implementation in python [https://github.com/abetlen
 This is the {config["repo"]}/{config["file"]} model.
 """
 
-examples = [
-    ["Tell me a joke about old houses.", "Why did the old house break up with the new house? Because it was too modern!"],
-    ["What is the square root of 64?", "The square root of 64 is 8."],
-    ["Insult me", ""],
-]
-
-gr.Interface(fn=generate_text, inputs=input_text, outputs=output_text, title="Llama Language Model", description=description, examples=examples).launch()
-
+gr.Interface(
+    fn=generate_text,
+    inputs=input_text,
+    outputs=output_text,
+    title="Llama Language Model",
+    description=description,
+).queue(max_size=16, concurrency_count=1).launch()