About the Token Counter
LLM providers bill by tokens, not characters or words, and token counts vary by tokenizer — a prompt that costs 500 tokens on one model can cost 650 on another. This tool estimates token counts for your prompt and projects per-request and monthly spend at a given traffic volume, so you can catch expensive prompts before they hit production.
Key Features
- Compare pricing across GPT-4o, GPT-4 Turbo, Claude Opus/Sonnet/Haiku, Gemini 1.5, and Llama 3.1.
- Projects monthly cost from a requests-per-day estimate.
- 100% client-side — your prompt text never leaves the browser.