AlloyDB now offers new AI functions and major performance improvements for running LLM calls directly within SQL queries. Three new functions are introduced: ai.summarize, ai.agg_summarize, and ai.analyze_sentiment, joining the existing GA functions (ai.generate, ai.rank, ai.if, ai.forecast). Two key performance breakthroughs are highlighted: Smart Batching, which intelligently groups AI function calls to reduce prompt overhead and achieve up to 2,400x throughput improvement; and Optimized AI Functions, which train a lightweight proxy model on your data to process decisions locally, achieving up to 100,000 rows/sec (23,000x improvement) and 6,000x cost reduction. Both features are currently in preview for ai.if and ai.rank.

7m read timeFrom cloud.google.com
Post cover image
Table of contents
The power of LLMs on your data: now significantly faster and cheaper
335 Impressions