Model overview
Two rate tables apply to one model, and that is what separates the Pro tier from the Flash rungs. Up to 200K tokens it bills $2 per million input, $0.20 cached and $12 output — a third above Flash on every line, and 6.7 times Lite's input with 4.8 times its output. Past 200K, Google publishes $4 input, $0.40 cached and $18 output, doubling the prompt rate and adding half again to generation for the request. Long agent runs therefore cross a step the cheaper rungs never meet. Since gemini-3-1-pro-preview lists the same reasoning, tools, multimodal input and context caching as the rest, the premium buys the tier itself, and the preview label is the other thing to weigh.
Billing conditions: Standard paid price for prompts up to 200K tokens. Above 200K, Google publishes $4 input, $0.40 cached input, and $18 output per million tokens.
- Context window
- 0 tokens
- Knowledge cutoff
- —
- Price last verified
- First observed
Descriptions, capabilities, and tariffs come from the linked source. First-party prices take priority; OpenRouter fallbacks are labelled.