AI Token Cost Calculator — GPT, Claude & Gemini
An AI token cost calculator works out what your app spends on AI each month. Pick a model, enter how many requests you make and how many tokens each request uses in and out, and it shows your monthly bill and cost per request — using the model's published per-token prices, which you can edit when prices change.
How it works
- Choose a model — GPT-4o, GPT-4o mini, Claude Opus 5, Claude Sonnet 4.6 or Gemini 2.5 Flash-Lite — or edit the input and output rates to match your provider.
- Enter your monthly requests and the average input and output tokens per request.
- See the monthly cost split into input and output, plus the cost of a single request.
Formula
Worked example
An app makes 10,000 requests a month, each using 1,000 input tokens and 500 output tokens. On GPT-4o ($2.50 per 1M input tokens, $10 per 1M output tokens) that is 10M × $2.50 = $25 for input plus 5M × $10 = $50 for output — $75 a month, or $0.0075 per request. The same workload on GPT-4o mini ($0.15/$0.60) costs about $4.50 a month.
Frequently asked questions
What is an AI token?
A token is a chunk of text — roughly three quarters of a word in English. A 1,000-token prompt is about 750 words. You pay for the tokens you send in and the tokens the model generates out.
How much does GPT-4o cost per 1,000 requests?
At 1,000 input tokens and 500 output tokens per request, 1,000 requests cost $7.50 on GPT-4o: $2.50 for 1M input tokens plus $5.00 for 0.5M output tokens. The same 1,000 requests cost about $0.45 on GPT-4o mini.
Why are output tokens more expensive than input tokens?
Generating text costs more compute than reading it, so every provider charges more per output token. Long answers are the expensive part of an AI bill — summarising and short replies keep costs down.
Which AI model is cheapest?
Of the presets here, Gemini 2.5 Flash-Lite ($0.10 per 1M input, $0.40 output) is cheapest, followed by GPT-4o mini ($0.15/$0.60). The cheapest model is not always the best choice — accuracy and reasoning matter too, so test before switching.
How can I cut my AI token bill?
Shorten prompts, use the smallest model that does the job, cache repeated prompts, and avoid asking for long outputs you do not need. Moving a simple task from GPT-4o to GPT-4o mini cuts the cost by about 94%.
Are these AI prices current?
The rates shown are the published list prices and they change often — that is why every rate in the calculator is editable. Check your provider's pricing page and update the rates before relying on a figure.