How much does the OpenAI, Claude or Gemini API cost?

Prices are charged per million tokens and differ for input and output. Across the 16 models compared here, input ranges from $0.15 to $10.00 per million tokens and output from $0.60 to $50.00. Your bill depends on how many requests you make and how long the input and output are, which is what this calculator works out.

A token is a chunk of text a model reads or writes — roughly three quarters of an English word, so about 1.35 tokens per word. You pay for both the tokens you send and the tokens the model returns.

How can I reduce AI API costs?

Three levers, cheapest first: cache the repeated part of your prompt, which providers bill at roughly a tenth of the normal input rate; use the batch API for work that can wait, which halves the price at OpenAI and Anthropic; and only then consider a smaller model. Shortening output also helps, since output tokens usually cost several times more than input.

Free calculator · 16 models

What will your AI feature actually cost?

Enter how much you send and receive, and compare the real monthly bill across OpenAI, Anthropic and Google. Every price is transcribed from the provider's own pricing page and dated below — no estimates, no affiliate links.

Start from

Requests per month

Input tokens per request

Output tokens per request

Input served from cache: 0%

A repeated system prompt or document is usually the cached part. Output is never cached.

Measure in

Batch API (50% off)For work that can wait — OpenAI and Anthropic both halve the price.

Compare

Cheapest option

$6.00/month

GPT-4o mini · OpenAI

The same workload costs $450.00 on Claude Fable 5.1 — 75.0× more.

25.0M tokens per month.

Monthly cost for 10,000 requests, cheapest first
Model	Provider	Input	Output	Per request	Per month
GPT-4o miniOlder, very cheap	OpenAI	$0.15/M	$0.60/M	$0.0006	$6.00
GPT-5.6 LunaCheapest frontier-family tier	OpenAI	$0.20/M	$1.20/M	$0.0010	$10.00
Gemini 3.5 Flash-Lite	Google	$0.30/M	$2.50/M	$0.0019	$18.50
Gemini 3.8 Flash	Google	$0.75/M	$3.75/M	$0.0034	$33.75
GPT-5.4 mini	OpenAI	$0.75/M	$4.50/M	$0.0037	$37.50
Claude Haiku 4.5200K context	Anthropic	$1.00/M	$5.00/M	$0.0045	$45.00
Gemini 3.5 Flash	Google	$1.50/M	$9.00/M	$0.0075	$75.00
Gemini 2.5 Pro≤200K context; higher above	Google	$1.25/M	$10.00/M	$0.0075	$75.00
Claude Sonnet 5	Anthropic	$2.00/M	$10.00/M	$0.0090	$90.00
GPT-5.6 Terra	OpenAI	$2.00/M	$12.00/M	$0.010	$100.00
Gemini 3.1 Pro (preview)≤200K context; higher above	Google	$2.00/M	$12.00/M	$0.010	$100.00
Claude Sonnet 4.6	Anthropic	$3.00/M	$15.00/M	$0.013	$135.00
GPT-5.6 Sol	OpenAI	$4.00/M	$20.00/M	$0.018	$180.00
Claude Opus 5	Anthropic	$5.00/M	$25.00/M	$0.022	$225.00
GPT-6 AstraFlagship	OpenAI	$10.00/M	$50.00/M	$0.045	$450.00
Claude Fable 5.1Most capable	Anthropic	$10.00/M	$50.00/M	$0.045	$450.00

Where these prices come from

Each provider's figures are transcribed from its own pricing page. Prices change without notice — check the source before committing budget.

OpenAI API pricingchecked 2026-09-07
Anthropic Claude pricingchecked 2026-09-07
Gemini API pricingchecked 2026-09-07

Text tokens on the standard short-context tier only. Long-context tiers, images, audio, fine-tuning, and reasoning-token overhead all cost more — treat this as a floor, not a quote. Found a stale number? Tell us.