Update app_quant_latent.py

app_quant_latent.py

@@ -232,6 +232,7 @@ try:
         transformer=transformer,
         text_encoder=text_encoder,
         torch_dtype=torch_dtype,
+        attn_implementation="kernels-community/vllm-flash-attn3",
     )
     pipe.to(device)
     log("✅ Pipeline built successfully.")