Google launches Gemini 3.1 Flash-Lite, prioritizing speed and affordability for high-volume AI workloads.

“Gemini 3.1 Flash-Lite reaches its first response token 2.5 times faster than Gemini 2.5 Flash and pushes out over 360 tokens per second.”
— The Decoder, Tech Analysis
2.5xSpeed improvement
sh.25/1MInput token price
.50/1MOutput token price
363 tok/sOutput speed