Google introduces Gemini 3.1 Flash-Lite, optimized for high-volume AI workloads

Google has announced Gemini 3.1 Flash-Lite, a new AI model designed for large-scale processing tasks. The company states the model outperforms Gemini 2.5 Flash across reasoning, reliability, and scalability while reducing operational costs. The release introduces adjustable thinking levels, enabling users to control computational overhead based on task complexity—from zero overhead for routine processing to full reasoning for complex scenarios.