Update app.py

app.py

@@ -7,7 +7,7 @@ import os
 
 @spaces.GPU
 def load_model(model_name):
-    return pipeline("text-generation", model=model_name, device_map="cuda", torch_dtype=torch.bfloat16, trust_remote_code=True, token=os.environ["token"])
+    return pipeline("text-generation", model=model_name, device_map="cuda", torch_dtype=torch.bfloat16, trust_remote_code=True, token=os.environ["token"], use_fast=True)
 @spaces.GPU()
 def generate(
     model_name,