Free tool · No sign-up

AI API Cost Calculator

Estimate what an AI feature will cost to run. Enter how many requests you expect and how long they are, and compare Claude, GPT and Gemini models side by side.

Quick start

Number of requests

Input tokens / request

Output tokens / request

Rough guide: 1,000 tokens is about 750 English words. Input includes your instructions, chat history and any documents you send.

Show in ₹ too (optional)

Requests per month: Based on 30 days.

Standard list prices in USD per 1 million tokens, checked on 21 Oct 2026. Prices change often, so confirm on the official pages: Anthropic, OpenAI, Google. Taxes, free tiers and volume discounts are not included. Every price above is editable.

How AI API pricing works

AI providers charge per token, a small chunk of text. You pay for the tokens you send (input: your instructions, the conversation so far and any documents) and the tokens the model writes back (output). Output tokens usually cost several times more than input tokens.

The formula is simple: (input tokens × input price + output tokens × output price) ÷ 1,000,000, multiplied by the number of requests.

Ways to lower your AI bill

Pick the smallest model that does the job well. Classification and simple replies rarely need the largest model. Test a few on your real data.
Keep prompts short. In chatbots, the whole conversation is sent again with every message, so long chats get expensive. Summarise or trim old messages.
Use prompt caching and batch processing where your provider offers them. Both can cut costs a lot for repeated instructions or work that doesn't need an instant answer.
Cap the output length so the model doesn't write more than you need.

More free tools

WhatsApp Link Generator Create a wa.me click-to-chat link and QR code with a pre-filled message. AI Prompt Template Library Ready-to-use prompts for marketing, sales, support, coding and operations.