Gemini API pricing (2026): what Argon and Flash cost now, and after the price rises
updated · prices checked on vendor pages
quick answer: Gemini 4 Argon is $2 in / $10 out per million tokens until november 30, 2026, then $4 / $20 from december 1. Gemini 3.8 Flash is $0.75 / $3.75 until december 31, 2026, then $1.50 / $7.50 from january 1, 2027. both double, a month apart, and Google prints both dates on its own pricing page. Flash has a free tier; Argon doesn't. the standard prices are on our daily price check, last confirmed 2026-10-22.
the Gemini API is #2 of 8 on our LLM APIs board (8.5/10): our pick for long-context and agentic work on a budget. the "on a budget" bit has an expiry date, so here's what you pay now, what you'll pay later, and how to soften the landing.
the price table
per million tokens, standard paid tier, in USD:
| model | input now | output now | from | input after | output after |
|---|---|---|---|---|---|
| Gemini 4 Argon | $2.00 | $10.00 | december 1, 2026 | $4.00 | $20.00 |
| Gemini 3.8 Flash | $0.75 | $3.75 | january 1, 2027 | $1.50 | $7.50 |
| Gemini 3.6 Flash | $0.75 | $3.75 | january 1, 2027 | $1.50 | $7.50 |
| Gemini 3.5 Flash-Lite | $0.30 | $2.50 | no change announced |
output prices include thinking tokens, so a model that thinks hard bills you for the thinking too. context caching follows the same schedule: Argon $0.10 per million cached tokens, then $0.20 from december 1 (implicit caching only); 3.8 Flash $0.075, then $0.15 from january 1, plus $0.50 per million tokens per hour of storage, then $1.00.
what a job actually costs
say an agent run reads 1 million tokens and writes 200,000 (output includes thinking). our sums, from the prices above:
| model | today | after the rise |
|---|---|---|
| Gemini 4 Argon | $4.00 | $8.00 (from december 1) |
| Gemini 3.8 Flash | $1.50 | $3.00 (from january 1) |
| Gemini 3.8 Flash, batch | $0.75 | $1.50 (from january 1) |
same work, twice the bill. if you've built a cost model on Argon's intro price, it's wrong from december 1.
the cheaper ways to run it
- batch: half price. Google's batch API is 50% off for jobs that can wait. Argon on batch is $1 / $5 now, then $2 / $10 from december 1. 3.8 Flash on batch is $0.375 / $1.875, then $0.75 / $3.75 from january 1. evals, backfills and overnight jobs belong here.
- flex: same price as batch. Argon and 3.8 Flash on flex match their batch prices.
- priority costs more. Argon on priority is $3.60 / $18, then $7.20 / $36 from december 1. only worth it if latency is the product.
- grunt work goes to Flash. route the easy calls (classify, extract, summarise) to Flash and save Argon for the hard ones. even after january it's under half Argon's december price.
- contributor tier: invite-only preview at the batch price, but your prompts and outputs are used to improve Google's products. fine for a hobby project, not for client code.
is the Gemini API free?
partly. the free tier covers the Flash models (3.8 Flash and 3.6 Flash are "free of charge" for input and output) and Google AI Studio itself is free to use. Gemini 4 Argon is not on the free tier. the catch: on the free tier your content is used to improve Google's products. on the paid tier it isn't.
the extras that show up on the bill
- grounding with Google search: 5,000 free search requests a month (shared across Gemini 3 and newer models), then $14 per 1,000. one prompt can fire several searches, and each one counts.
- grounding with Google maps: same deal, 5,000 a month free, then $14 per 1,000.
- code execution and URL context: billed as normal tokens at the model's rate.
how it stacks up
| rank | tool | best for | free tier | paid from | numptr score |
|---|---|---|---|---|---|
| #2 | Google Gemini API | long-context and agentic tasks on a budget | free tier on Flash models (Argon not on free tier) | Gemini 4 Argon $2/$10 per M tokens through nov 30, 2026, then $4/$20 | 8.5/10 |
| #1 | Anthropic Claude API | coding agents, tool use | small amount of free credits for new API users | Sonnet 5.5 $2/$10 per M tokens | 9.0/10 |
| #3 | OpenAI API | general-purpose and multimodal apps | unknown | GPT-6.1 Sol $2/$10 per M tokens | 8.5/10 |
| #5 | DeepSeek API | cheap agent and coding workloads on a tight token budget | unknown: no free tier stated (the pricing page mentions a granted balance) | deepseek-flash $0.30/$1.20 per M tokens at peak, $0.15/$0.60 off-peak | 8.0/10 |
| #6 | MiniMax API | cheap agent and coding workloads that need a long context | unknown: no free tier stated on the pricing page | MiniMax-M3.1-Flash $0.15/$0.60 per M tokens, flat rate | 8.0/10 |
on price per token, Argon today sits level with Sonnet 5.5 and GPT-6.1 Sol at $2 / $10. from december 1 it's $4 / $20, which is Opus 5.5 money. DeepSeek and MiniMax are far cheaper per token if you can live with them, and OpenRouter lets you switch between the lot on one key. the Claude API is still #1 on our board for coding agents and tool use.
which should you use?
- long context, agentic work, and you're building before december: Argon, and enjoy it while it lasts
- high volume, cost first: 3.8 Flash, on batch where the job can wait
- prototyping for free: the free tier on Flash, as long as you're fine with Google using the data
- coding agents and tool use, price second: the Claude API
- cheapest tokens going: DeepSeek off-peak or MiniMax
more on the coding side: best ai coding agent 2026 and the ai coding agent price sheet. every LLM API we rank, with a reason: the LLM APIs board.
faq
how much does the Gemini API cost?
Gemini 4 Argon is $2 per million input tokens and $10 per million output tokens until november 30, 2026, then $4 and $20 from december 1. Gemini 3.8 Flash is $0.75 and $3.75 until december 31, 2026, then $1.50 and $7.50 from january 1, 2027. batch is half price.
when does Gemini API pricing go up?
Gemini 4 Argon doubles on december 1, 2026. Gemini 3.8 Flash and 3.6 Flash double on january 1, 2027. context caching on those models doubles on the same dates.
is the Gemini API free?
there's a free tier on the Flash models, and Google AI Studio is free to use. Gemini 4 Argon is paid only. on the free tier, Google uses your content to improve its products.
how much is Gemini 4 Argon?
$2 in / $10 out per million tokens through november 30, 2026, then $4 / $20. on batch it's $1 / $5, then $2 / $10.
is the Gemini API cheaper than Claude?
today, Argon and Claude Sonnet 5.5 are the same price at $2 / $10 per million tokens. from december 1 Argon costs $4 / $20, the same as Claude Opus 5.5. Flash is cheaper than both, before and after january.
tools in this guide
sources
checked 2026-10-21
- Gemini developer API pricing (Argon, 3.8 Flash, 3.6 Flash, 3.5 Flash-Lite; Standard, batch, flex, priority and contributor tiers; free tier; grounding): https://ai.google.dev/gemini-api/docs/pricing
- comparison table prices (from our tools data, re-checked daily): https://platform.claude.com/docs/en/about-claude/pricing ; https://developers.openai.com/api/docs/pricing ; https://api-docs.deepseek.com/quick_start/pricing ; https://platform.minimax.io/docs/pricing/overview