Google Cloud has outlined how to use AlloyDB AI Functions with Gemini models to improve search indexing for Chinese, Japanese and Korean text in AlloyDB, targeting full-text and hybrid search workloads in managed PostgreSQL environments. Standard PostgreSQL text parsing assumes words are separated by spaces, creating difficulties for logographic languages written as continuous text, where a database may treat an entire sentence as a single lexeme. AlloyDB can now call Gemini models directly from SQL through in-database AI Functions, allowing developers to apply word segmentation and stop-word removal before creating full-text indexes. The proposed design keeps raw text, segmented text, search vectors and embeddings in the same table, with generated columns ensuring automatic refresh of the full-text search vector and vector embedding when segmented text is updated. For large document collections, Google Cloud described a batching process built with a PL/pgSQL stored procedure that groups rows into arrays, submits a batched call to Gemini, unpacks outputs with GENERATE_SERIES, updates target records and commits each batch immediately. One example uses the Chinese sentence "你们研究所有十个图书馆"; without segmentation, PostgreSQL may index the full string as a single unit, but after preprocessing with Gemini, terms such as "研究所" and "图书馆" become searchable. Google Cloud also outlined that for databases containing only Chinese text, developers may prefer PostgreSQL's simple configuration, while for bilingual datasets mixing Chinese with English, the english configuration may be more practical.