Update app.py

app.py

@@ -14,9 +14,8 @@ chunk_size = int(os.environ.get("CHUNK_SIZE", 128))
 default_max_characters = int(os.environ.get("DEFAULT_MAX_CHARACTERS", 258))
 
 model = SentenceTransformer(model_name)
-# model.to(device="cuda")
+model.to(device="cuda")
 
-@spaces.GPU
 def embed(queries, chunks) -> dict[str, list[tuple[str, float]]]:
     query_embeddings = model.encode(queries, prompt_name="query")
     document_embeddings = model.encode(chunks)