Google Cloud Surfaces Gemini 3.2 Flash Lite Model with Inference Costs 95% Lower Than GPT-5.5

According to Beating.AI monitoring, a new model option named gemini-3.2-flash-lite-live-preview has appeared in Google Cloud’s model selection list as of May 17. The “lite” and “live” suffixes indicate Google is creating a specialized version optimized for ultra-low-latency real-time interactions.

Abacus.AI CEO Bindu Reddy previously disclosed that Gemini 3.2 Flash achieves 92% of GPT-5.5’s coding and reasoning capabilities while keeping inference costs at just 1/20th of GPT-5.5’s, with most queries returning responses below 200 milliseconds. Industry observers expect this cost-optimized lightweight model to be formally unveiled at Google I/O on May 20.