What does each Gemini model cost per million tokens?
Per Google's Gemini API pricing page, all prices are in USD per million tokens, standard tier, text input.
| Model | Input | Output (including thinking) | Context caching | Free tier |
|---|---|---|---|---|
| Gemini 4 Carbon | $2 through Nov 30, 2026; $4 from Dec 1 | $10 through Nov 30; $20 from Dec 1 | $0.10 through Nov 30; $0.20 from Dec 1 | Not available |
| Gemini 4 Argon | $2 through Nov 30, 2026; $4 from Dec 1 | $10 through Nov 30; $20 from Dec 1 | $0.10 through Nov 30; $0.20 from Dec 1 | Not available |
| Gemini 4 Neon | $3 | $15 | $0.15 | Not available |
| Gemini 4 Hydrogen | $1.50 | $7.50 | $0.075 | Free of charge |
| Gemini 3.8 Flash | $0.75 through Dec 31, 2026; $1.50 from Jan 1, 2027 | $3.75 through Dec 31; $7.50 from Jan 1 | $0.075 through Dec 31; $0.15 from Jan 1 | Free of charge |
| Gemini 3.5 Flash-Lite | $0.30 (text, image, video, audio) | $2.50 | $0.03 | Free of charge |
As reviewed on November 19, 2026. Google also lists other models (image, speech, embedding, robotics and earlier Gemini versions) at their own rates, and says prices can differ on its enterprise platform. Check the page before you budget.
How do batch, caching and search change the price?
Each is a separate line on Google's pricing page.
- Batch API: Google lists a 50 percent cost reduction on the Paid tier.
- Context caching: cached input is billed at the caching rate in the table, which is a fraction of the input price.
- Thinking tokens: Google says every token the model processes is counted and billed, including thinking tokens, so output prices apply to the thinking too.
- Grounding with Google Search: 5,000 free requests a month shared across Gemini 3 and newer models, then $14 per 1,000 requests, with a default cap of 1,500 requests a day per project.
- Documents: Google says PDF tokens are billed at the image token rate.
Which prices change soon?
Google's API pricing page dates two price increases for specific models.
For help building with the API and setting spend limits, see our AI implementation page. To get a key and see the tier caps, read the API key guide.
- December 1, 2026: Gemini 4 Carbon and Argon input, output and caching prices double ($2 to $4 input, $10 to $20 output).
- January 1, 2027: Gemini 3.8 Flash prices double ($0.75 to $1.50 input, $3.75 to $7.50 output).
- What to do: if you are budgeting a project, use the later rate. This advice is ours, not Google's.
Frequently asked questions
How much does the Gemini API cost?
Google lists paid-tier input and output rates per million tokens of $0.30 and $2.50 for Gemini 3.5 Flash-Lite, $1.50 and $7.50 for Gemini 4 Hydrogen, $3 and $15 for Gemini 4 Neon, and $2 and $10 for Gemini 4 Carbon and Argon until November 30, 2026.
Is there a batch discount?
Yes. Google lists the Batch API at a 50 percent cost reduction on the Paid tier.
When do Gemini 4 Carbon prices go up?
Google's page lists Gemini 4 Carbon and Argon at $4 input and $20 output per million tokens starting December 1, 2026, double the earlier $2 and $10.
Does the API charge for thinking tokens?
Yes. Google says every token the model processes is counted and billed, including thinking tokens.
Can you build with the Gemini API for us?
Yes. We scope what to build, set spend limits and keep keys safe. Book a call to start.
Need help with Gemini?
We help small and mid-size teams choose, set up and run AI tools. See our Gemini guides and services, and the AI implementation, and the Claude pricing guides.




