Gemini Pricing Guide

Gemini API pricing varies significantly between models, making model selection one of the most effective ways to control cost.

How the estimate is calculated

For example, Google currently lists paid Gemini 2.5 Flash text, image, and video input at $0.30 per million tokens and output, including thinking tokens, at $2.50 per million tokens.

Gemini 2.5 Pro uses a different structure. Google currently lists input at $1.25 per million tokens for prompts up to 200,000 tokens and $2.50 above 200,000 tokens. Output pricing is listed at $10 per million tokens for prompts up to 200K and $15 per million tokens above that threshold.

What can change the total

This means cost comparisons require more than asking, “How much does Gemini cost?”

You need to know:

Plan with a realistic usage example

Which Gemini model is being used How many input tokens are processed How many output or thinking tokens are generated Whether prompts cross long-context thresholds Whether caching is used Whether grounding or other paid features are enabled

High-volume classification or extraction tasks may favor a lower-cost model, while difficult reasoning tasks may justify a more capable model.

Practical checks before you proceed

Calculate cost per successful task, not merely price per token.

Pricing changes over time and newer Gemini generations may have completely different rates.

Practical checks before you proceed

Last reviewed: August 2026. Confirm current Google AI pricing before making production budgets or purchasing decisions.