Upload server.py with huggingface_hub

server.py

@@ -22,8 +22,8 @@ def load_model():
         "zai-org/GLM-Image",
         torch_dtype=torch.bfloat16,
     )
-    pipe.enable_sequential_cpu_offload()
-    print("[image] Model loaded with sequential CPU offload!", flush=True)
+    pipe.enable_model_cpu_offload()
+    print("[image] Model loaded with model CPU offload!", flush=True)
 
 @app.route("/health", methods=["GET"])
 def health():