Fix ZeroGPU startup

app.py

@@ -7,6 +7,8 @@ from dataclasses import dataclass
 from pathlib import Path
 from typing import Any
 
+import spaces  # Must be imported before torch-related libraries on ZeroGPU
+
 import faiss
 import gradio as gr
 import numpy as np
@@ -127,11 +129,13 @@ if index.d != index_config["embedding_dimension"]:
 
 # Load retrieval models once when the app starts.
 embedding_model = SentenceTransformer(
-    EMBEDDING_MODEL_NAME
+    EMBEDDING_MODEL_NAME,
+    device="cpu",
 )
 
 reranker = CrossEncoder(
-    RERANKER_MODEL_NAME
+    RERANKER_MODEL_NAME,
+    device="cpu",
 )
 
 
@@ -412,6 +416,7 @@ def enrich_candidates(
     return enriched_results
 
 
+@spaces.GPU(duration=60)
 def rerank_candidates(
     query: str,
     candidates: list[dict[str, Any]],