🔵 Gemini Token Calculator
What a prompt costs on Google's Gemini models. Several Gemini prices are promotional and have a published end date, so the figure you budget with today may not be the figure you pay in two months.
Gemini's promotional pricing, and the dates it expires
Google's pricing has a feature none of the other providers share to the same extent: several current rates are explicitly promotional, with a published date on which they rise. Budgeting against today's Gemini price without noting that date is a reliable way to be wrong by 100% in two months.
Current prices
| Model | Input | Output | Cache read |
|---|---|---|---|
| Gemini 4 Argon | $2 | $10 | — |
| Gemini 3.8 Flash | $0.75 | $3.75 | — |
| Gemini 3.5 Flash-Lite | $0.3 | $2.5 | — |
| Gemini 3.1 Pro | $2 | $12 | — |
| Gemini Omni Flash | $1.5 | $9 | — |
What is about to change
Gemini 4 Argon is $2 input and $10 output until 30 November 2026, after which it doubles to $4 and $20. Gemini 3.8 Flash is $0.75 and $3.75 until 31 December 2026, then doubles to $1.50 and $7.50. Both are flagged on this page, and the calculator shows what your volume would cost after the change as well as before it.
The free tier is real but limited
Several Gemini models have a genuine free tier with no payment required, which is unusual among the major providers and makes Gemini the cheapest way to prototype. The limits are on request rate and daily volume rather than on capability, so the free tier is suitable for development and unsuitable for production.
Multimodal pricing is simpler here
Gemini 3.5 Flash-Lite charges one input rate for text, image, video and audio alike, which makes multimodal cost estimation unusually straightforward — most providers price each modality separately or do not support them at all. Gemini Omni Flash goes further and prices video output per token, working out at roughly $0.10 per second of 720p.
The 200K prompt threshold on Pro
Gemini 3.1 Pro costs $2 input and $12 output up to a 200,000-token prompt, and $4 and $18 above it. As with Anthropic's Haiku threshold, the higher rate applies to the whole request, so a long-context application should measure where its prompts actually land rather than assuming.
Prices come from Gemini API pricing and were checked on 2026-10-26. AI pricing changes frequently — confirm against the provider before committing to a budget.
Worth pairing with the ai token calculator, which uses the same inputs for a different question, and with the chatgpt token cost for the step after.
Frequently Asked Questions (FAQs)
How much does the Gemini API cost?
On 2026-10-26 it ranged from $0.25 per million input tokens on the Flash-Lite models to $2 on Gemini 4 Argon, with output between $1.50 and $12. Several of those are promotional rates.
Which Gemini prices are about to rise?
Gemini 4 Argon doubles to $4 and $20 on 1 December 2026, and Gemini 3.8 Flash doubles to $1.50 and $7.50 on 1 January 2027. Both were promotional launch prices.
Does Gemini have a free tier?
Yes, on several models including 3.8 Flash and the Flash-Lite family. The limits are on request rate and daily volume rather than capability, which makes it good for prototyping and unsuitable for production traffic.
How does Gemini price images and video?
Flash-Lite charges a single input rate covering text, image, video and audio. Omni Flash prices video output per token, working out at about $0.10 per second of 720p. That is simpler than most providers, which price each modality separately.
What is the 200K threshold on Gemini Pro?
Prompts up to 200,000 tokens cost $2 input and $12 output; above that it becomes $4 and $18, applied to the whole request. Long-context applications should measure where their prompts actually fall.
Is Gemini cheaper than GPT or Claude?
At the light end they are close, and the ranking depends entirely on your input-to-output ratio. Compare on your own volume in the pricing comparison rather than on headline rates.
Are these prices current?
They were read from Google's Gemini API pricing page on 2026-10-26. Given that several are promotional with published end dates, recheck before any budget longer than a quarter.