Update app.py

app.py

@@ -1,5 +1,5 @@
 import spaces
-import gradio as gr
+# import gradio as gr
 
 from vid2persona import init
 from vid2persona.pipeline import vlm
@@ -27,7 +27,7 @@ async def extract_traits(video_path):
         gr.Button(interactive=True)
     ]
 
-@spaces.GPU
+# @spaces.GPU
 async def conversation(
     message: str, messages: list, traits: dict,
     model_id: str, max_input_token_length: int, 
@@ -42,7 +42,7 @@ async def conversation(
         prompt_tpl_path, model_id, 
         max_input_token_length, max_new_tokens,
         temperature, top_p, top_k, 
-        repetition_penalty, hf_token=None#init.hf_access_token
+        repetition_penalty, hf_token=init.hf_access_token
     ):
         last_message = messages[-1]
         last_message[1] = last_message[1] + partial_response
@@ -69,7 +69,7 @@ async def regen_conversation(
             prompt_tpl_path, model_id, 
             max_input_token_length, max_new_tokens,
             temperature, top_p, top_k, 
-            repetition_penalty, hf_token=None#init.hf_access_token
+            repetition_penalty, hf_token=init.hf_access_token
         ):
             last_message = messages[-1]
             last_message[1] = last_message[1] + partial_response