Do I need an API key?

No. This tool parses a response you already received. Everything runs locally in your browser and nothing is uploaded.

What are cache_creation and cache_read tokens?

cache_creation_input_tokens are the tokens written into the prompt cache (billed at a premium), while cache_read_input_tokens are read back from the cache on later calls (billed at a deep discount). They make repeated long prompts much cheaper.

How is the cost estimated?

It applies published per-million-token list prices for the selected model. Cache writes are billed at 1.25x the input rate and cache reads at 0.1x, which the estimate reflects.

Where is the usage object in an Anthropic response?

At the top level of a Messages API response there is a usage object containing input_tokens, output_tokens, and the two cache fields. The parser reads it directly.

Anthropic Usage Stats Parser

Parse Claude token usage and estimate cost

The Anthropic Usage Stats Parser reads the usage object from a Claude Messages API response and breaks out every token category Anthropic bills separately: regular input, output, cache-creation (writes), and cache-read (hits). It then estimates the dollar cost using the correct rate for each category. No API key is required — you are parsing a response you already have.

How it works

The tool parses your JSON and reads usage.input_tokens, usage.output_tokens, usage.cache_creation_input_tokens, and usage.cache_read_input_tokens. Anthropic bills these at different rates: cache writes cost 1.25× the base input price, while cache reads cost only 0.1×. The estimate multiplies each bucket by the right per-million-token price for the model you select, then sums them. Everything runs locally.

Tips and notes

A healthy prompt-caching setup shows large cache_read_input_tokens and small cache_creation_input_tokens after the first call — that is where the savings come from. If you see cache creation on every request, your cache breakpoints or TTL may be wrong. Use this alongside the OpenAI usage parser when comparing providers so you are reading the same kind of numbers. Nothing you paste leaves your browser.