Gemini 2.5 Pro pricing
Google's GA flagship. 2M context window (largest in the industry) with dual-tier pricing - watch for the long-context surcharge above 200K input.
Input
$1.25/ 1M tok
Output
$10.00/ 1M tok
- Context window
- 2M
- Max output
- -
- Cached input
- $0.125 / 1M
- Verified
- 2026-04-06
Long-context tier
Above 200K input tokens, this model bills at $2.50 input / $15.00 output per 1M tokens.
Gemini 2.5 Pro is Google's GA flagship and the only model we track with a 2M-token context window. Standard pricing is $1.25 per 1M input and $10 per 1M output, which looks aggressive - until your prompt crosses 200,000 input tokens and the rate doubles to $2.50 / $15. Long documents and large contexts land in the surcharge tier routinely, so it's worth planning around.
Cached input at $0.125 per 1M (a 90% discount on standard) makes repeated prefixes cheap, but Google's context caching is explicit - you create cache objects and reuse them via the API - so wire it up deliberately. For workloads that don't need the 2M context, Gemini 2.5 Flash is 4x cheaper on input.
Calcis counts input tokens against Google's own countTokens API and automatically applies the long-context surcharge in the estimate whenever your prompt crosses the 200K threshold.
Estimate your cost on Gemini 2.5 Pro
Paste your prompt into the estimator, pick Gemini 2.5 Pro, and see the exact dollar cost - input tokens counted with the provider's own tokenizer, output tokens predicted by our regression model.
Frequently asked
- When does Gemini 2.5 Pro long-context pricing kick in?
- At 200,000 input tokens. Below that, $1.25 input / $10 output per 1M. Above, $2.50 input / $15 output per 1M. The surcharge applies to the whole request once you cross the threshold.
- How much does Gemini 2.5 Pro cost per request?
- A 1,000-token prompt with a 500-token reply costs about $0.006. A 300K-token prompt (long-context tier) with a 1K reply costs about $0.77. Always check which tier your prompt falls into.
- Is Gemini 2.5 Pro cheaper than GPT-5?
- On input, roughly the same ($1.25 vs $1.25 per 1M). On output, Gemini is marginally more expensive ($10 vs $10 - tie at the headline rate). Pick on context size: Gemini's 2M vs GPT-5's 400K is a clear differentiator for long-document work.
- What tokenizer does Gemini 2.5 Pro use?
- Google's proprietary SentencePiece variant. Calcis uses Google's countTokens API for the authoritative count - the same boundary Google uses to bill.
Pricing verified 2026-04-06 from the provider's rate card.