Google · Direct API

Gemini 3.5 Flash-Lite pricing

Google publishes a $0.02 batch cache-read rate rather than exactly half the standard cache rate.

DirectOfficial provider

Input$0.3per 1M tokens

Cached input$0.03per 1M tokens

Output$2.5per 1M tokens

Context1.05Mmaximum window

This is a first-party rate checked against the model provider's official pricing page.

Caching

2,000 input + 500 output tokens$0.00185 / request