tps field

app.py

@@ -2,7 +2,7 @@ from llama_cpp import Llama
 from huggingface_hub import hf_hub_download
 import gradio as gr
 from typing import Tuple, List
-import time  # Import the time module
+import time
 
 DESCRIPTION = f"""
 # Chat with Arco 500M as GGUF on CPU
@@ -36,26 +36,26 @@ def predict(message: str, history: List[List[str]], max_new_tokens: int = DEFAUL
 
     # Initialize reply for this round
     reply = ""
-    
+
     # Initialize token count and start time
     token_count = 0
     start_time = time.time()
-    
+
     # This will produce a generator of output chunks
     stream = pipe(
-        prompt,  
+        prompt,
         max_tokens=max_new_tokens,
         stop=["</s>"],
         stream=True
     )
-    
+
     # Send each token stream output to the user
     for output in stream:
         new_text = output['choices'][0]['text']
         reply += new_text
         token_count += len(new_text.split()) # Estimate tokens by counting spaces
         history[-1][1] = reply # Update the current reply in history
-        
+    
         # Calculate elapsed time and TPS
         elapsed_time = time.time() - start_time
         if elapsed_time > 0:
@@ -81,6 +81,6 @@ with gr.Blocks() as demo:
         label="Max New Tokens",
     )
     status_field = gr.Text(label="Status", interactive=False, visible=True) # Add Status field
-    textbox.submit(predict, [textbox, chatbot, max_new_tokens_slider], [textbox, chatbot], progress=status_field)
+    textbox.submit(predict, [textbox, chatbot, max_new_tokens_slider], [textbox, chatbot], )
 
-demo.queue().launch()
+demo.queue().launch()