Gemini 3.1 Flash-Lite enters preview at $0.25 per million input tokens

Google introduced Gemini 3.1 Flash-Lite on 3 March in preview through AI Studio’s Gemini API and Vertex AI. Announced prices were $0.25 per million input tokens and $1.50 per million output tokens. The company aimed it at repeated workloads such as translation, content moderation and structured interface generation.

Thinking levels give developers another way to control the work spent on an individual request. Google also cited faster time to first answer token and output speed than Gemini 2.5 Flash, using reported benchmark results. Those comparisons should not substitute for measuring the actual prompt mix. At high request volumes, input length, generated output and the amount of reasoning can all affect whether a nominally inexpensive model produces an inexpensive application.