Skip to main content
anonpaste/tools

Prompt cost estimator

Estimate usage with editable model pricing.

Prompt cost & context estimator

Estimate costs using your own rates. Defaults are illustrative USD examples, not provider quotes. Select the tokenizer your model uses and enter its context limit.

Tokenizer profile
Input tokens0Exact token count
Input cost$0.00
Output cost$0.008000
Estimated total$0.008000
Context remaining127,000
Context usage0.78%
1,000 of 128,000 tokens

Rates and context limits are assumptions you supply. This excludes caching discounts, batch pricing, reasoning tokens, tool schemas, attachments, taxes and request-template overhead. Exact text tokenization does not guarantee an exact API bill. No API requests are made.

Your text is processed only in this browser and is not uploaded.

About Prompt cost estimator

Work out an approximate model usage cost from token counts and editable rates. Enter the rates you have verified for your provider and the expected output, rather than assuming the tool has live prices.

Steps, and limitations

How to use it

  1. Set the input and output rates and the usage assumptions.
  2. Enter or estimate the relevant token counts.
  3. Review the calculated cost and revise the assumptions for other scenarios.

Things to know

  • This is not a live price feed or a billing statement. Cached usage, batches, requests, and other charges can affect the real cost.
  • Verify current rates with the provider before making a spending decision.

Related tools

All in this category

Navigation

Find a tool

esc

Favorites, recent, and quick access

Image compressorReduce image file size with adjustable quality.
JSON formatterFormat, validate, and minify JSON data.
Merge PDFCombine multiple PDF files into one document.
Password generatorGenerate random passwords with configurable character sets.
↑↓ navigate↵ open toolNothing you type here is uploaded.

Preferences

These preferences are saved only in this browser.

Tool input stays in memory and is cleared when you leave the page. Downloaded files are managed by your browser. AI model caches, when used, are separate from these preferences.