Type anything below. It splits into tokens live, right in your browser — no login, nothing saved.
It's what you pay for
AI tools are usually priced per token — both what you type in and what the AI writes back. Longer prompts and longer answers cost more.
It's why AI "forgets"
Every conversation has a token limit. Once you're near it, the AI starts losing track of things you said earlier — that's the token limit at work.
It's not the same as word count
A rough rule of thumb: 1 token ≈ 4 characters in English. Code, jargon, and other languages often use more tokens per word than plain English does.
What are tokens? A plain guide
How LLMs count and charge for text, from words to context limits.
LLM context windows, explained
What a context window is and why bigger isn't always better.
Token budget: planning & execution
How to plan a token budget for agents and long workflows.
Try the Context Window Visualizer →
See how far your text fills up a model's context window.
It's the small chunk of text an AI model actually reads — usually a piece of a word, not a whole word. In English that's roughly 3-4 letters per token. This matters because you're charged and limited by token count, not by how many words or characters you typed.
The model learned its vocabulary from huge amounts of text, so common words got their own single token while rarer words, typos, and non-English text get chopped into several smaller pieces. Try the "Non-English text" example above to see this clearly.
It switches between two ways OpenAI's models read text — an older one (used by GPT-4 and GPT-3.5) and a newer, more efficient one (used by GPT-4o and beyond). Same idea, slightly different chunking. Most people can leave it on the default.
Roughly, not exactly. Anthropic (the company behind Claude) uses its own counting method that isn't public, so this tool uses OpenAI's public one as the closest stand-in. Counts across different AI companies are usually within 10-20% of each other for normal English text.
No. Everything happens right in your browser — nothing you type is sent to a server, logged, or stored anywhere. No account or login required.
Context Window Visualizer
See how much of a model's context window your text fills up, and what happens when you run out of room.
FROG in a Bowl Prompt Builder
Fill in Format, Role, Objective, Goal, and Context — get a copy-ready structured prompt in seconds.
Machine Learning Types
Three tiny games show what supervised, unsupervised, and reinforcement learning actually mean.
Neural Network Playground
Drag two sliders and watch a real, tiny neural network turn them into a decision, live.
Generative AI Playground
Watch AI write one word at a time by predicting what's most likely to come next.
RAG Playground
Ask a question, watch notes get retrieved, then see a grounded answer versus a hallucination.
Embedding Map
Click two words and see why similar meanings sit near each other — the idea behind vector search.
Attention Visualizer
Click a word and see which others a toy transformer looks at — including the classic “it” puzzle.
Prompt Injection Lab
Watch a pasted email try to override a support agent, then flip a switch that treats it as data.
