Update app.py

app.py

@@ -34,20 +34,20 @@ pipe.to("cuda")
 # pipe.transformer.set_attention_backend("flash_hub")
 pipe.transformer.set_attention_backend("_flash_3_hub")
 
-@spaces.GPU(duration=1500)
-def compile_transformer():
-    with spaces.aoti_capture(pipe.transformer) as call:
-        pipe("arbitrary example prompt")
+# @spaces.GPU(duration=1500)
+# def compile_transformer():
+#     with spaces.aoti_capture(pipe.transformer) as call:
+#         pipe("arbitrary example prompt")
 
-    exported = torch.export.export(
-        pipe.transformer,
-        args=call.args,
-        kwargs=call.kwargs,
-    )
-    return spaces.aoti_compile(exported)
+#     exported = torch.export.export(
+#         pipe.transformer,
+#         args=call.args,
+#         kwargs=call.kwargs,
+#     )
+#     return spaces.aoti_compile(exported)
 
-compiled_transformer = compile_transformer()
-spaces.aoti_apply(compiled_transformer, pipe.transformer)
+# compiled_transformer = compile_transformer()
+# spaces.aoti_apply(compiled_transformer, pipe.transformer)
 
 # ---------------------------------------------------------------------------
 # Generation