Update app.py

app.py

@@ -22,7 +22,7 @@ MODEL_NAME = MODEL_ID.split("/")[-1]
 
 TITLE = "<h1><center>VL-Chatbox</center></h1>"
 
-DESCRIPTION = f'<h3><center>MODEL: <a href="https://hf.co/models/{MODEL_ID}">{MODEL_NAME}</a></center></h3>'
+DESCRIPTION = f'<h3><center>MODEL: <a href="https://hf.co/{MODEL_ID}">{MODEL_NAME}</a></center></h3>'
 
 CSS = """
 .duplicate-button {
@@ -65,12 +65,12 @@ def stream_chat(message, history: list, temperature: float, max_new_tokens: int)
             conversation.append({"role": "user", "content": message['text']})    
     print(f"Conversation is -\n{conversation}")
 
-    streamer = TextIteratorStreamer(tokenizer, **{"skip_special_tokens": True, "skip_prompt": True, 'clean_up_tokenization_spaces':False,}) 
+  #  streamer = TextIteratorStreamer(tokenizer, **{"skip_special_tokens": True, "skip_prompt": True, 'clean_up_tokenization_spaces':False,}) 
 
     generate_kwargs = dict(
         image=image,
         msgs=conversation,
-        streamer=streamer,
+#        streamer=streamer,
         max_new_tokens=max_new_tokens,
         temperature=temperature,
         sampling=True,
@@ -79,7 +79,7 @@ def stream_chat(message, history: list, temperature: float, max_new_tokens: int)
     if temperature == 0:
         generate_kwargs["sampling"] = False
     
-
+"""
     thread = Thread(target=model.chat, kwargs=generate_kwargs)
     thread.start()
 
@@ -87,6 +87,9 @@ def stream_chat(message, history: list, temperature: float, max_new_tokens: int)
     for new_text in streamer:
         buffer += new_text
         yield buffer
+"""
+    response = model.chat(**generate_kwargs)
+    return response
 
 
 chatbot = gr.Chatbot(height=450)