What is a token, and why is AI so cheap?
The whole thing, in one breath
A token is a small chunk of text, roughly three-quarters of a word. AI companies count tokens going in and out and charge a tiny amount per thousand. That is why a whole scene usually costs pennies, not dollars.
Like you're five
Imagine the AI reads and writes with LEGO bricks instead of whole words. "Cat" might be one brick. "Unbelievable" might be three. When you send a message, the company counts the bricks you handed over and the bricks it handed back, and charges a sliver of a cent per brick. Since bricks are cheap, even a big castle of a chapter comes out to spare change.
A little more grown-up
In and out are both counted
You pay for the tokens you send (your prompt plus all the story context) and the tokens the model writes back. A long chapter with a lot of world detail uses more bricks than a quick one-line edit. That is the whole billing model, and it is why bigger requests cost a little more.
The numbers are genuinely small
Prices are quoted per million tokens and they are low. A typical scene, context and all, might run a few thousand tokens. Multiply that by a per-million rate and you land at cents. This is not marketing softness. It is just the math of paying real cost with your own key.
AIStoryHub tries not to waste your bricks
Sending the whole novel with every request would burn tokens for nothing, so the studio is careful about what it packs in: the characters actually in the scene, the relevant lore, the recent chapters. You get the context that matters without paying to re-send the parts that do not.
So what, for your writing?
Once you know AI is billed by the brick, the fear of a surprise bill goes away. You can watch your usage, set limits with your provider, and write freely knowing a normal session costs less than a stick of gum. Cheap fuel is the whole point of bringing your own key.
New to keys and cost? Start on the free model with zero tokens to worry about.
Start freeWords that showed up
- Token
- A small chunk of text, about three-quarters of a word, that AI is measured and billed in.
- Context window
- The most tokens a model can hold at once. Its short-term memory for a single request.
- Usage
- How many tokens you have spent. Your provider shows this on your account dashboard.