Gemini 3.1 Pro token & cost calculator

Gemini 3.1 Pro is Google's current flagship Gemini model — the high-capability Pro tier, carrying a 1,000,000-token context window and Google's signature two-tier pricing: $2 input / $12 output per million at or below 200,000 input tokens, stepping up to $4 / $18 for any request that crosses that threshold. It supersedes Gemini 2.5 Pro, which stays available at a lower rate.

The tier boundary is the thing to plan around. At 200K tokens — one-fifth of the way into the context window — the whole request re-rates to roughly double the input price. So Gemini 3.1 Pro rewards keeping prompts lean and punishes sprawling context in a way flat-priced models like GPT-4.1 don't. If your workload lives above 200K, model the upper tier as your real rate.

Expected response (output tokens)

Prompt

Client-side. Never uploaded.

0 / 1,000,000 charactersContext window: 1,000,000 tokens

Or start with an example

Total estimated cost

$0.012Gemini 3.1 Pro

Tokens±3% approx

Input cost

$0.00

Output cost (est.)

$0.012

@ 1,024 response tokens

Context used

of 1,000,000

Verified 2026-07-06 · ±3%

Saved scenariosnone yet

Saved on this browser only — never uploaded. Up to 10 scenarios.

Tip: save a scenario when you have a prompt + model + response length you might revisit. Useful for sizing features before committing to a vendor.

Verify privacysince this page loaded — updates live

Prompt uploads0Always 0 — by design

Outgoing requests0Analytics + page assets only — no prompt content

Cookies on this origin0Vercel Analytics + Clarity may set first-party cookies

localStorage keys0Theme preference + saved scenarios live here

Server endpoints1/api/og only — accepts title + subtitle, never prompt text

Inspect

Open DevTools → Network. Type into the calculator. No request bodies should contain your prompt text.

Pricing

Gemini 3.1 Pro is tier-priced on input size. The calculator picks the right tier automatically; the breakdown is below.

Tier	Input $/M	Output $/M
≤ 200,000 input tokens	$2	$12
All input	$4	$18
Context window	1,000,000 tokens

Verified against ai.google.dev on 2026-07-06.

Worked examples

Watch the tier boundary: the 150K example is billed at $2/$12, but the 300K and 900K examples cross 200K and re-rate to $4/$18. Paste your own prompt above to price it exactly.

Scenario	Input	Output	Cost
Short chat turn A typical Q&A turn — well within the low tier.	800	400	<$0.01
150K input (low tier) A large prompt still under the 200K threshold: $2/$12.	150,000	2,000	$0.324
300K input (high tier) Crosses 200K — the whole request re-rates to $4/$18.	300,000	3,000	$1.25
Near-full context Deep in the upper tier, close to the 1M ceiling.	900,000	4,000	$3.67

The jump between the 150K and 300K rows is the tier surcharge made visible — the extra 150K of input more than doubles the input bill because the rate itself changes, not just the token count. For sustained large-context work, Gemini 3.1 Pro's upper tier is often where a flat-priced million-token model like GPT-4.1 wins on cost.

How is this counted?

Google doesn't publish a client-side Gemini tokenizer, so we approximate with o200k_base (via js-tiktoken), which empirically tracks Gemini within a few percent on English prose and source code. For exact, billable counts, Google's countTokens API endpoint returns the precise figure before you send a request. Inputs over 50,000 characters run in a Web Worker. Treat this as a budgeting estimate — typical drift is ~3%.

FAQ

Why does the price jump on large prompts?

Gemini 3.1 Pro is tier-priced on input size. At or below 200,000 input tokens it's $2 input / $12 output per million; above 200,000 tokens the whole request is billed at $4 input / $18 output — roughly double the input rate. The calculator applies the right tier automatically based on what you paste, so a 150K prompt and a 250K prompt are priced on different rates.

How is Gemini 3.1 Pro different from Gemini 2.5 Pro?

3.1 Pro is the current flagship; 2.5 Pro is the previous generation (still available). Both carry a 1,000,000-token window and both tier at 200K, but 3.1 Pro costs more — $2/$12 (rising to $4/$18 above 200K) versus 2.5 Pro's $1.25/$10 (rising to $2.50/$15). Pay the step up when you need the newer model's capability; 2.5 Pro remains the cheaper option for the same context length.

How accurate is the token count?

Approximate. Google doesn't publish a client-side Gemini tokenizer, so we estimate with o200k_base, which tracks Gemini within a few percent on English prose and code. For exact billing counts, Google's countTokens API endpoint returns the precise figure before you send. Treat this calculator as a budgeting tool — typical drift is ~3%.

What is the context window?

1,000,000 tokens of input. The tier surcharge kicks in at 200,000 input tokens, well below the ceiling, so most large-context requests are billed at the higher tier — worth modeling before you commit to a long-context design.

Does the tier surcharge apply to output too?

Yes. Above 200,000 input tokens the whole request moves to the upper tier: input goes from $2 to $4 and output from $12 to $18 per million. The threshold is measured on input size, but it re-rates both sides of the bill.

Compare against every other model

To see this exact prompt scored against every supported model, sorted by total cost, paste it into the home calculator and toggle Compare across all models. The comparison is most revealing on prompts above 200K tokens, where Gemini's tier surcharge separates it from flat-priced long-context models.

The natural comparison set: Gemini 2.5 Pro (the previous-generation Pro, cheaper at the same context length), Gemini 2.5 Flash (the budget Gemini tier), and GPT-4.1 (the flat-priced million-token alternative with no tier surcharge).