diff --git a/scripts/rag_build_query.py b/scripts/rag_build_query.py
index bdf878d..bc28d29 100644
--- a/scripts/rag_build_query.py
+++ b/scripts/rag_build_query.py
@@ -36,7 +36,7 @@ logger.setLevel(logging.INFO)
 
 
 # BGEEmbedding 类，继承自 HuggingFaceEmbedding，用于生成查询的嵌入
-class BGEEmbedding(HuggingFaceEmbedding):
+class BGEEmbedding(HuggingFaceEmbedding): 
     def _get_query_embedding(self, query: str) -> List[float]:
         try:
             # 在查询前加上前缀，生成嵌入向量
@@ -47,6 +47,9 @@ class BGEEmbedding(HuggingFaceEmbedding):
             # 转换为 float32 类型
             embedding = np.array(embedding, dtype=np.float32)
 
+            # 将 numpy 数组转换为列表，确保 embedding 是一个列表
+            embedding = embedding.tolist()
+
             # 使用 logger 打印数据类型
             logger.info(f"Query embedding dtype after conversion: {embedding.dtype}")
             return embedding
@@ -61,8 +64,8 @@ class BGEEmbedding(HuggingFaceEmbedding):
             prefix = "Represent this sentence for searching relevant passages: "
             embeddings = super()._get_query_embeddings([prefix + q for q in queries])
 
-            # 转换为 float32 类型
-            embeddings = [np.array(embedding, dtype=np.float32) for embedding in embeddings]
+            # 转换为 float32 类型并转换为列表
+            embeddings = [np.array(embedding, dtype=np.float32).tolist() for embedding in embeddings]
 
             # 使用 logger 打印数据类型
             logger.info(f"Batch query embeddings dtype after conversion: {embeddings[0].dtype}")