AI Token Counter
Count exact LLM tokens for GPT, Claude and Gemini prompts.
About the AI Token Counter
Tokenize any prompt with the real BPE tables used by GPT-4o, GPT-4 and embedding models (o200k_base, cl100k_base, p50k_base). See token counts, tokens-per-word, context-window usage and input cost, plus raw token IDs — all computed offline in your browser.
Examples
Short prompt
Summarise this article in three bullets.Output
9 tokens (o200k_base)Keyboard shortcuts
- Copy the main outputCtrl / ⌘ + Shift + C
- Download the resultCtrl / ⌘ + S
- Share this toolCtrl / ⌘ + Shift + S
- Reset the inputsAlt + R
- Open the tool search paletteCtrl / ⌘ + K
Related tools
LLM API Cost Calculator
AI & LLM
Estimate GPT, Claude and Gemini API spend per call and per month.
LLM Context Window Calculator
AI & LLM
See if your system, context and user messages fit the model window.
AI Dataset Deduplicator
AI & LLM
Remove exact and near-duplicate rows from fine-tuning datasets.
AI Dataset Statistics Analyzer
AI & LLM
Profile a JSONL training set: token stats, roles, duplicates and cost.
AI PII & Secret Redactor
AI & LLM
Strip emails, keys, tokens and personal data before prompting an LLM.
AI Tool Schema Builder
AI & LLM
Build function-calling schemas for OpenAI, Claude, Gemini, MCP and the AI SDK.
Frequently asked questions
Read more
- How LLM tokens really work (and why your word count is a bad estimate)
BPE splits words, punctuation and whitespace into pieces. Here is how tokens are counted, why context windows overflow, and how to plan a prompt that fits.
Version 1.0.0 · Updated 2026-08-14 · Runs entirely in your browser