Skip to content
Yunedia Tools

Calculadora de custos de API

Estime os custos da API de LLM a partir do número de tokens de entrada e saída, usando uma tabela de preços datada retirada das páginas oficiais de preços.

Cached tokens are part of the input tokens and are billed at the cached rate when the model has one.

Per request
$0.040
Per day
$4.00
Per month (×30)
$120.00

Prices as of 2026-11-16. Sources: OpenAI · Anthropic · Google

Como usar

  1. Choose a model.
  2. Enter input and output token counts (and cached input tokens if you know them), or paste text to estimate.
  3. Set requests per day to see the daily and monthly (×30) totals.

Observações

Prices are the list prices verified on the providers' official pricing pages on 2026-11-16 (USD per 1M tokens). Discounts, free tiers, batch pricing and exchange rates are not included. Text mode estimates tokens as characters ÷ 4, which is not an exact count. The data date and sources are always shown inside the tool.

Perguntas frequentes

What are cached input tokens?
The part of your input that a prompt cache reuses. Many models bill reused input at a lower rate. Here you can mark part of the input as cached; it is billed at the model's cached rate, or at the normal input rate when the model has no cached price.
Will this match my actual bill?
It is a list-price estimate. Actual bills vary with your plan, discounts, minimums and how tokens are metered. Always confirm on the provider's pricing page and your billing console.