Update app.py

app.py

@@ -35,13 +35,7 @@ def predict(message, history):
         temperature=0.7,
         num_beams=1
     )
-    t = Thread(target=model.generate, kwargs=generate_kwargs)
-    t.start()
-    partial_message = ""
-    for new_token in streamer:
-        partial_message += new_token
-        if '<|im_end|>' in partial_message:
-            break
-        yield partial_message
+    output = model.generate(**generate_kwargs)
+    partial_message = streamer.decode(output[0], skip_special_tokens=True)
 
 gr.ChatInterface(predict).launch()