Token Counter
Count tokens, words, and characters for GPT, Claude, and Gemini models with real-time API cost estimates.
Estimate LLM tokens and API spend before you ship a prompt
Paste draft prompts, system messages, or RAG context and see approximate token counts plus input/output cost ranges for popular models. Counts use character-to-token ratios; pair this with your provider's tokenizer when invoices must be exact.
Related tools
For character-level stats without model pricing, try the Word Counter. To compare JSON payload size, use the JSON Formatter.
Frequently Asked Questions
How accurate are the token counts?
They are estimates based on average characters per token for each model family. Code, non-English text, and special symbols can tokenize differently than plain English prose.
Is my text uploaded to a server?
No. Counting and cost math run entirely in your browser.