Gemini 3.1 Flash-Lite: Built for intelligence at scale

Google DeepMind · 4 months ago

Google released Gemini 3.1 Flash-Lite, a faster and cheaper model in its Gemini 3 series designed for high-volume developer workloads. The model costs $0.25 per million input tokens and $1.50 per million output tokens, with 2.5X faster response times and 45% higher output speed than its predecessor while achieving an Elo score of 1432 on Arena.ai benchmarks. Developers can now build cost-effective real-time applications with adjustable reasoning levels for tasks ranging from translation and content moderation to complex UI generation and instruction-following.