How Gemini API pricing works
A Gemini request costs input tokens times the input price, plus output tokens times the output price, with cached input billed at a discount. What changes the bill:
- Input: everything you send on each call: system instructions, chat history, documents, images and tool results. Audio input costs more on Flash models.
- Output: the answer plus the model's thinking tokens, billed at the same output price.
- Long prompts: Pro models charge more per token when a prompt is over 200K tokens: Gemini 3.1 Pro goes from $2 to $4 input and from $12 to $18 output.
- Context caching: cached input costs 10% of the input price, plus a storage fee per hour for explicit caches.
- Batch: 50% off input and output for asynchronous work.
Gemini counts about 4 characters per token in English. Paste a real prompt into the token counter to see what it costs per call and per month.
What each Gemini model costs for the same job
Take 1 million classification requests a month, each with 1,000 input tokens and a 20-token answer, at today's prices:
| Model | Input cost | Output cost | Monthly total |
|---|---|---|---|
| Gemini 3.1 Pro Preview | $2,000 | $240 | $2,240 |
| Gemini 4 Argon | $2,000 | $200 | $2,200 (doubles to $4,400 on Dec 1) |
| Gemini 3.8 Flash | $750 | $75 | $825 (doubles to $1,650 on Jan 1) |
| Gemini 3.5 Flash-Lite | $300 | $50 | $350 |
| Gemini 3.1 Flash-Lite | $250 | $30 | $280 |
| Gemini 2.5 Flash-Lite | $100 | $8 | $108 |
Thinking tokens are billed as output, so a reasoning model that thinks for 2,000 tokens before a 20-token answer pays for 2,020 output tokens. Set a thinking budget on simple tasks, or use a Flash-Lite model that thinks less. For routing and the other levers, see the LLM cost optimization guide.
Gemini API free tier
Gemini 3.8 Flash, the Flash-Lite models and the Gemini 2.5 models can be used for free in Google AI Studio and the API, within rate limits. Gemini 4 Argon and Gemini 3.1 Pro are paid only. On the free tier, Google may use your prompts and responses to improve its products; on the paid tier it does not. The free tier suits prototypes, not production traffic or private data.
Context caching and batch prices
Gemini caches context two ways. Implicit caching is automatic: when a request repeats the start of a recent one, the repeated tokens are billed at the cached price with nothing to set up, but hits are not guaranteed. Explicit caching stores a prefix you create for a set time, with guaranteed hits, and charges storage per million tokens per hour: $4.50 on Pro models, $1 on Flash-Lite, $0.50 on Gemini 3.8 Flash. Gemini 4 Argon supports implicit caching only.
Explicit caching pays off when a large prefix is read often. A 100K-token cache on Gemini 3.1 Pro costs $0.45 an hour to store, and each request that reads it saves 100K × ($2 − $0.20) / 1M = $0.18. From 3 requests an hour, the cache is cheaper.
The Batch API halves input and output prices for asynchronous jobs: Gemini 3.1 Pro drops to $1 input and $6 output.
Other charges on the Gemini API
- Grounding with Google Search: 5,000 free search requests a month, shared across Gemini 3 and newer models, then $14 per 1,000. Gemini 2.5 models get 1,500 free requests a day, then $35 per 1,000 grounded prompts.
- Scheduled increases: Gemini 4 Argon goes to $4 input and $20 output on December 1, 2026; Gemini 3.8 Flash to $1.50 and $7.50 on January 1, 2027. Budget next quarter at the new prices.
- Audio input: Flash models charge more for audio than for text, images and video, for example $0.50 against $0.25 per million tokens on Gemini 3.1 Flash-Lite.
Older Gemini models
Still available on the API at these prices. Newer models in the same tier are usually as cheap or cheaper for better results.
| Model | Input | Output | Cached input | Batch input | Batch output |
|---|---|---|---|---|---|
Gemini 2.5 Pro Prompts over 200K tokens: $2.50 input, $15 output | $1.25 | $10.00 | $0.125 | $0.625 | $5.00 |
Gemini 2.5 Flash | $0.30 | $2.50 | $0.03 | $0.15 | $1.25 |
Gemini 2.5 Flash-Lite | $0.10 | $0.40 | $0.01 | $0.05 | $0.20 |
Frequently asked questions
How much does the Gemini API cost?
Gemini API prices run from $0.10 per million input tokens and $0.40 per million output tokens on Gemini 2.5 Flash-Lite to $2 and $12 on Gemini 3.1 Pro (prompts up to 200K tokens). Gemini 4 Argon costs $2 and $10 until November 30, 2026, then $4 and $20. Cached input costs 10% of the input price, and the Batch API halves every price.
Is the Gemini API free?
Partly. Gemini 3.8 Flash, the Flash-Lite models and the Gemini 2.5 models have a free tier with rate limits, where Google may use your data to improve its products. Gemini 4 Argon and Gemini 3.1 Pro are paid only.
What is the cheapest Gemini API model?
Gemini 2.5 Flash-Lite, at $0.10 per million input tokens and $0.40 per million output tokens, half that with the Batch API. It also has a free tier with rate limits. Among newer models, Gemini 3.1 Flash-Lite costs $0.25 input and $1.50 output.
How much is Gemini 3.1 Pro per million tokens?
$2 per million input tokens and $12 per million output tokens for prompts up to 200K tokens, and $4 and $18 above that. Cached input costs $0.20 ($0.40 over 200K), and the Batch API halves input and output prices.
Will Gemini 4 Argon get more expensive?
Yes. Google lists Gemini 4 Argon at $2 input and $10 output per million tokens through November 30, 2026, and $4 and $20 from December 1, 2026. Cached input goes from $0.10 to $0.20.
Does Gemini charge for thinking tokens?
Yes. Thinking tokens are billed as output tokens at the model's output price. A thinking budget limits them on tasks that do not need long reasoning.
What is the difference between implicit and explicit caching on Gemini?
Implicit caching is automatic: repeated prompt prefixes are billed at the cached price when the cache happens to hit, with no storage fee. Explicit caching is a cache you create, with guaranteed hits and a storage fee per million tokens per hour.
Is Gemini cheaper than Claude?
It depends on the tier. Gemini 4 Argon and Claude Sonnet 5.5 both cost $2 input and $10 output until December 1, when Argon doubles. At the small end, Claude Haiku 5.5 costs $0.10 and $0.50, close to Gemini 2.5 Flash-Lite at $0.10 and $0.40. See Claude API pricing for the full comparison.