AI model cost calculator
Calculate the expected cost of your AI prompts and API requests in real time. Adjust the number of input, output and cached tokens easily.
Enter parameters
example prices — illustrative, check current rates with the provider yourself
Quick select:
Rate per 1,000,000 tokens (€)
Estimated cost
Total estimated cost
€ 0,423
Exact: € 0.423000
Input cost (100.000 tokens)
€ 0,2350
Cached input cost (0 tokens)
€ 0,0000
Output cost (20.000 tokens)
€ 0,1880
Calculation formula & explanation
Costs are computed from the entered rates per million tokens:
Kosten = (InputTokens / 1M × PrijsInput) + (CachedTokens / 1M × PrijsCached) + (OutputTokens / 1M × PrijsOutput)
Note: actual costs on your invoice can differ slightly through exchange rate fluctuations (USD to EUR) and rounding per API call.
Background & theory
Want to understand exactly how tokenization works, what context windows mean and how prompt caching can lower your input costs by as much as 90% on modern models (around 50% on older ones)?
Read the article: Token pricing models explained →LLMNet API routing
Ready to get started with a unified API gateway for multiple LLM providers? See the documentation and endpoints on the main site.
Visit api.llmnet.nl →
