SourceUnifying Structured and Unstructured Data Insights with BQ Search Innovations
These features reduce the need for complex external pipelines when analyzing documents like PDFs. This allows businesses to perform natural language searches over unstructured data more cost-effectively.
- General Availability of Autonomous Embedding Generation supporting text and images via Vertex AI and Gemma models
- General Availability of AI.SEARCH with up to 133x slot efficiency gain for single-query execution
- Public Preview of Hybrid Search uniting semantic vector search and BM25/RRF lexical matching
- Integrated document analytics capabilities including AI.PARSE, AI.CHUNK_DOC, and BigQuery Graph