Gemini 4 Argon goes from $2.00 input and $10.00 output to $4.00 input and $20.00 output per million tokens on December 1, 2026. Gemini 3.8 Flash and 3.6 Flash go from $0.75 input and $3.75 output to $1.50 input and $7.50 output on January 1, 2027. Cached input, Batch and Flex double with them. Calls before each date keep the old price.
What changes and when
Every Standard rate on these three models doubles: input, output (including thinking tokens) and cached input. Gemini 3.8 Flash and 3.6 Flash cache storage also goes from $0.50 to $1.00 per million tokens per hour. Prices are for the paid tier of the Gemini Developer API.
| Model | From | Input per 1M | Cached input per 1M | Output per 1M |
|---|---|---|---|---|
| Gemini 4 Argon | December 1, 2026 | $2.00 → $4.00 | $0.10 → $0.20 | $10.00 → $20.00 |
| Gemini 3.8 Flash | January 1, 2027 | $0.75 → $1.50 | $0.075 → $0.15 | $3.75 → $7.50 |
| Gemini 3.6 Flash | January 1, 2027 | $0.75 → $1.50 | $0.075 → $0.15 | $3.75 → $7.50 |
We track every scheduled change and shutdown across OpenAI, Anthropic and Google on the LLM API price change tracker, with an RSS feed.
What it does to each request
Doubling a price doubles every request on that model. Per 1,000 requests at Standard rates:
| Request | Gemini 4 Argon now | From December 1, 2026 | Gemini 3.8 Flash now | From January 1, 2027 |
|---|---|---|---|---|
| Chat reply (3,000 input tokens, 500 output) | $11.00 | $22.00 | $4.13 | $8.25 |
| Document summary (30,000 input tokens, 1,500 output) | $75.00 | $150 | $28.13 | $56.25 |
| Agent task (80,000 input tokens (60,000 cached), 6,000 output) | $106 | $212 | $42.00 | $84.00 |
A product that runs 100,000 agent tasks a month on Gemini 4 Argon goes from $10,600 to $21,200. A support bot answering a million chat replies on Gemini 3.8 Flash goes from $4,125 to $8,250.
How Gemini compares after the increase
Today Gemini 4 Argon costs the same per token as GPT-6.1 Sol ($2.00 input and $10.00 output) and Claude Sonnet 5.5 ($2.00 input and $10.00 output). From December 1, 2026 it costs the same as Claude Opus 5.5 ($4.00 input and $20.00 output).
Gemini 3.8 Flash costs less than GPT-6.1 Terra ($1.00 input and $5.00 output) today. From January 1, 2027 it costs more. Claude Haiku 5.5 ($0.10 input and $0.50 output) and Gemini 3.5 Flash-Lite ($0.30 input and $2.50 output) cost less than either.
Per-token prices are only part of it: tokenizers, cache discounts and long-prompt rates differ. See the side-by-side pages: Claude Sonnet 5.5 vs Gemini 4 Argon, GPT-6.1 Sol vs Gemini 4 Argon, Gemini 3.8 Flash vs GPT-6.1 Terra and Gemini 3.8 Flash vs 3.5 Flash-Lite.
Five things to do before the date
- 01Find which customers and features use these models
The increase only costs you where the volume is. Rank workflows by Gemini 4 Argon and 3.8 Flash spend before deciding anything else.
- 02Move work that can wait to Batch or Flex
Both stay 50% off after the increase. From December 1, 2026, Gemini 4 Argon through Batch costs $2.00 input and $10.00 output: what Standard costs today.
- 03Route simple requests to a cheaper model
Classification, extraction and short replies often run fine on Gemini 3.5 Flash-Lite ($0.30 input and $2.50 output, no change announced). Test quality on your own prompts first.
- 04Reuse more of each prompt
Cached input still costs a fraction of the input price after the increase. Long, stable system prompts and documents are the cheapest place to save.
- 05Review fixed-price plans and credits
Customers on flat plans keep paying the same while their Gemini cost doubles. Check which of them stay profitable at the new price before renewals.
See the impact per customer
Your Google bill will show the total going up. It won’t show which customers caused it. Ganivra prices every Gemini call at the rate in effect that day, including the old and new prices on either side of the change, and adds it up by customer, feature and workflow next to what each customer pays you. Setup is two lines with the Python SDK or REST API.
Gemini price increase: frequently asked questions
When does the Gemini API price increase?
Gemini 4 Argon prices double on December 1, 2026. Gemini 3.8 Flash and Gemini 3.6 Flash prices double on January 1, 2027. Google's pricing page lists the old price "through" the day before and the new price "starting" on that date.
How much will Gemini 4 Argon cost?
$4.00 input and $20.00 output per million tokens from December 1, 2026, up from $2.00 input and $10.00 output. Cached input goes from $0.10 to $0.20.
How much will Gemini 3.8 Flash cost?
$1.50 input and $7.50 output per million tokens from January 1, 2027, up from $0.75 input and $3.75 output. Cached input goes from $0.075 to $0.15, and cache storage from $0.50 to $1.00 per million tokens per hour.
Do calls made before the increase cost the new price?
No. Each call is billed at the price in effect on the day it ran, so calls before the change date keep the old price.
Are other Gemini models going up?
Google's pricing page announces increases for Gemini 4 Argon, 3.8 Flash and 3.6 Flash only. Gemini 3.5 Flash-Lite stays at $0.30 input and $2.50 output per million tokens, with no change announced.
Does the Batch API price go up too?
Yes. Batch and Flex stay 50% off, so they double along with the Standard price. From December 1, 2026, Gemini 4 Argon through Batch costs $2.00 input and $10.00 output, the same as its Standard price today.
Received a November 19 billing notice? Follow the Gemini API prepay migration guide.
Before the increase lands
Know which customers a price change hits.
Start a 14-day Ganivra trial and see your Gemini, OpenAI and Claude cost per customer. No credit card required.