LLM Cost Calculator - AI API Pricing
Find out how much each AI API request costs and your estimated monthly spend.
Cost per request
$0,0440
$0,0200
Input
$0,0240
Output
Estimated monthly cost (1000 requests)
$44,00
Prices per million tokens in USD, verified on official pricing pages (August 2026). DeepSeek prices are reference and may change. Use "Enter prices manually" if your provider has changed rates.
Other calculators
Frequently asked questions
How is AI API cost calculated?
AI APIs charge per token: each model has a price per million input tokens and another per million output tokens. The cost of one request is (input tokens x input price) + (output tokens x output price), divided by one million.
What is prompt caching?
It reuses input tokens already processed in previous requests, with a significant discount (up to 90% on some providers). If your app always sends the same system prompt, caching cuts your bill dramatically.
How many tokens does normal text have?
As a reference, 1,000 tokens equals roughly 750 English words. A page of text is usually 500-1,000 tokens.
Which AI API is the cheapest?
Among the major providers, DeepSeek V3 ($0.27/1M input) and GPT-5.6 Luna ($0.20/1M) are among the most affordable. Small models like Haiku or Flash offer the best cost/speed balance for simple tasks.
How to Calculate AI API Costs
The math is straightforward: multiply your input tokens by the model input price, your output tokens by the output price, and divide by one million. For example, with GPT-5.6 Terra ($2/1M input, $12/1M output), a request with 10,000 input tokens and 2,000 output tokens costs 0.02 + 0.024 = $0.044. At 1,000 requests per month, that is $44.
Tips to Reduce Your AI Bill
Use prompt caching for repeated tokens, pick the smallest model that handles your task, cap output length (max_tokens), and measure your real token usage before scaling. For batch workloads, most providers offer 50% discounts.