Update app.py

app.py

@@ -18,8 +18,11 @@ DEFAULT_INFERENCE_STEPS = 1
 dtype = torch.float16
 pipe = FLUXPipelineWithIntermediateOutputs.from_pretrained(
     "ostris/OpenFLUX.1", torch_dtype=dtype
-).to("cuda")
-pipe.vae = AutoencoderTiny.from_pretrained("madebyollin/taef1", torch_dtype=torch.float16).to("cuda")
+)
+pipe.vae = AutoencoderTiny.from_pretrained("madebyollin/taef1", torch_dtype=torch.float16)
+pipe.load_lora_weights("ostris/OpenFLUX.1", weight_name="openflux1-v0.1.0-fast-lora.safetensors", adapter_name="fast")
+pipe.set_adapters("fast")
+pipe.to("cuda")
 # pipe.transformer.to(memory_format=torch.channels_last)
 # pipe.transformer = torch.compile(
 #     pipe.transformer, mode="max-autotune", fullgraph=True