~/devtools / llm-cost
tool::llm-cost

LLM Cost
Calculator

Enter input/output token counts to compare API costs across models in real-time. Powered by live OpenRouter data.

ad · 728×90
/
Loading data from OpenRouter...
ad · 300×250
// about this tool

About this tool

LLM API costs can vary dramatically between providers and models — sometimes by 10x or more. When building production applications that rely on language models, even small differences in per-token pricing can translate into significant monthly expenses as usage scales up. This calculator lets you enter your expected token usage and instantly compare costs across hundreds of models side by side.

Pricing data is fetched in real time from the OpenRouter public API. OpenRouter aggregates models from OpenAI, Anthropic, Google, Meta, and other major providers under a single unified API, and maintains up-to-date pricing information. No API key is required — the data is pulled from a public endpoint every time you load the page.

Use the 1K/1M token toggle to estimate costs at different scales, from small development tests to large-scale production workloads. The provider filter lets you narrow results to a specific vendor, and the cheapest model is highlighted in green so you can spot it at a glance. All cost calculations happen client-side in your browser.

Use Cases
  • ▸Estimating monthly API costs for production LLM applications
  • ▸Comparing models to find the best cost-performance balance
  • ▸Benchmarking OpenAI vs Anthropic vs Google pricing
  • ▸Finding the optimal model for specific input/output token ratios
  • ▸Budget planning for side projects and enterprise-scale deployments
FAQ
Q. How often is the pricing data updated?
Pricing data is fetched live from the OpenRouter API each time you open the page. There is no caching — you always see the most current prices available.
Q. What is the difference between input and output tokens?
Input tokens (prompt) are the text you send to the model, while output tokens (completion) are the text the model generates in response. Most models charge 2–3x more for output tokens than input tokens.
Q. Are models listed as Free really free?
Some models on OpenRouter have a price of $0. However, they may have rate limits or other conditions that could change. Always verify the current terms on the official OpenRouter website.