Free tool

GPT & Gemini API Cost Calculator

Pick a real task — write an article, build a game, a month of coding — then switch between GPT and Gemini models to compare official provider rates with your price at a flat 50% off. No sign-up, no card.

Full-time development with an AI coding agent — all day, every day, ~22 workdays.

Flat −50% off official prices on every request — see how billing works
You pay for a month of coding
$11.63
$23.25−50%

for a month of coding on GPT-6.1 Terra · official OpenAI price minus your 50% discount

You save$11.63≈ ×2 more work per $
Get API key

Same OpenAI-compatible API, models and responses. You just pay less per call.

Every GPT model for this task

Official OpenAI price vs your apiToken.sale price at −50%, for a month of coding.

ModelInput / 1MOutput / 1MOfficialYour priceYou save
GPT-6.1 Terra
gpt-6.1-terra
$1$5$23.25$11.63$11.63
GPT-6.1 Sol
gpt-6.1-sol
$2$10$46.50$23.25$23.25
GPT-6 Astra
gpt-6-astra
$10$50$272.50$136.25$136.25
GPT-6 Sol
gpt-6-sol
$2$10$54.50$27.25$27.25
GPT-6 LunaCheapest
gpt-6-luna
$0.1$0.5$2.73$1.36$1.36
GPT-5.6 Sol
gpt-5.6-sol
$4$20$109.00$54.50$54.50
GPT-5.6 Terra
gpt-5.6-terra
$2$12$59.50$29.75$29.75
GPT-5.6 Luna
gpt-5.6-luna
$0.2$1.2$5.95$2.98$2.98

Each task uses a realistic total number of tokens to finish the whole job — expand “Adjust the tokens yourself” to tune it. GPT rows use standard rates from the pinned catalog. GPT-5.6 cache writes use 1.25× input. A single request over 272K input tokens has a long-context premium not inferred from these whole-task totals. Estimates only — your real bill depends on exact token usage.

How it works

Price a whole task, not a token

The calculator uses the pinned official OpenAI and Google rate catalogs. Nothing is sent anywhere — it all runs in your browser.

01

Pick a real task

Choose a whole job — write an article, build a game, analyze 500 memecoins, a month of coding — not a single request.

02

See the token budget

Each task carries a realistic total of input and output tokens to finish it end to end. Adjust them yourself if you like.

03

Compare the price

Switch between GPT and Gemini to see the official cost next to your apiToken.sale cost, with the cheapest model highlighted.

FAQ

GPT and Gemini API pricing, explained

How much do the GPT and Gemini APIs cost?+
OpenAI and Google bill per token. Official rates per 1M tokens include: GPT-6.1 Terra — $1 input / $5 output; GPT-6.1 Sol and GPT-6 Sol — $2 / $10; GPT-6 Astra — $10 / $50; GPT-6 Luna — $0.10 / $0.50; Gemini 4 Hydrogen — $1.50 / $7.50. Promotional rates for GPT-5.6 Sol, Gemini 4 Carbon, Gemini 4 Argon and Gemini 3.8 Flash change on fixed dates. Enter your token counts above to get an exact estimate for each model.
How is API pricing calculated?+
Cost = (input tokens ÷ 1,000,000 × input rate) + (output tokens ÷ 1,000,000 × output rate), plus provider-specific cache charges. Use the switch to apply the correct OpenAI or Google rates across every supported model.
How can I make the GPT and Gemini APIs cheaper?+
apiToken.sale serves the official GPT and Gemini models on one key and one balance, billed at a flat 50% below official prices on every request. Point Codex, Claude Code, Cursor, Cline or the OpenAI SDK at our endpoint — same models, same responses, half the price per call.
What is a token?+
A token is roughly ¾ of a word in English — about 4 characters. 1,000 tokens is ~750 words. Both the text you send (input) and the text the model generates (output) are counted, and output tokens are billed at a higher rate than input.
Are these prices exact?+
The rates come from the pinned official OpenAI and Google price catalogs, and the discounted column is your real apiToken.sale price. Your actual bill still depends on the exact token buckets and context size of each request, so treat whole-task totals as close estimates.