Free tool

LLM API cost calculator

Enter what one request sends and receives, and how many you make a month. Every model is priced from current list prices, cheapest first.

Loading current prices…

ModelPer requestPer month

Every host, with history

The calculator uses one price per model listing. Our LLM token prices dataset tracks the same models on OpenAI, Anthropic, Google, AWS Bedrock, Azure and other hosts, with every price change since we started, through the API or as a SQLite download.

Questions

Where do the prices come from?

OpenRouter's public model list, which states each model's price per million input and output tokens. We collect it every 6 hours into our LLM token prices dataset, which also tracks the same models on OpenAI, Anthropic, Google, AWS Bedrock, Azure and other hosts.

How is the cost worked out?

Cost per request = input tokens × input price + output tokens × output price, each per million tokens. The cached share of your input is priced at the cached-read price where the model has one, and at the normal input price where it does not. Monthly cost multiplies that by your requests per month.

Why does the same model cost different amounts elsewhere?

Each host sets its own price, and batch or regional versions are priced on their own. The calculator shows one price per model listing; the full dataset has every host, with history.

Does it include taxes, fine-tuning or images?

No. It covers text tokens only. Image, audio and video generation are priced per unit and are not included.

Start with 50 free calls

Create an account and verify your email: your own API keys and a free trial of 50 calls or 14 days, whichever ends first, no card needed. Then buy a credit pack from $15 or pick a plan from $49 a month.