Automation and AI

Token

Also called Tokens, token count

The small chunk of text a language model reads and writes — a word, part of a word, or a punctuation mark.

Quick facts: Token

Category
Automation and AI
Also called
Tokens, token count
Level
Intermediate
Affects
Cost per request, context limits, response length, speed
Where to see it
Provider usage dashboards, tokenizer pages, API billing screens
In this article4
  1. What a token measures
  2. Why tokens matter
  3. Common mistakes with tokens
  4. How to act on it

What a token measures

A language model does not read letters or whole words. Before anything happens, your text is split into tokens: common words usually become a single token, rarer or longer ones get broken into pieces, and punctuation and spacing count too. The model then works entirely in those pieces, predicting the next one over and over until the answer is finished.

Two consequences follow, and both cost money. Every request is charged by how many tokens go in and how many come out, so a long document pasted in and a long answer requested are both expensive. And every model has a ceiling on how many tokens it can hold in one conversation — its context window. Reach that ceiling and something has to be dropped, which is usually the earliest part of the conversation.

Why tokens matter

They turn vague worries about AI cost into something you can plan. If you know roughly how much text goes into a task and how much comes back, you can estimate what running it across a catalogue or an inbox will cost before you commit to it, rather than discovering it on an invoice.

They also explain behaviour that otherwise looks like a fault. An assistant that seems to forget an instruction given earlier in a long chat has not malfunctioned; the instruction has fallen outside what it can still see. An answer that stops mid-sentence has hit an output limit. A tool that gets slower and dearer as the conversation grows is doing exactly what it should, because every reply re-reads everything before it.

There is a point that matters in Nepal and anywhere multilingual. Languages written in scripts less represented in a model’s training tend to be split into more tokens per word than English. The same message in Nepali can cost noticeably more to process than its English translation, and it eats the context window faster.

Common mistakes with tokens

Confusing tokens with words when budgeting is the routine one, and it always errs in the wrong direction: the token count is higher than the word count, more so for names, product codes, URLs and non-English text. The second is pasting whole documents in when a relevant extract would do, then paying for the same background material on every follow-up question.

The third is letting one long-running conversation carry everything. Once the earliest part starts falling out of view, the model’s answers change without warning, and a chat that ran for hours quietly stops following the rules it was given at the start.

How to act on it

Send only what the task needs. Extract the relevant section rather than attaching the whole file, ask for the length of answer you actually want, and start a fresh conversation for a new job instead of continuing an old one out of convenience.

For anything you will run repeatedly, measure one instance first — check the usage figures in your provider’s dashboard — and multiply by realistic volume before rolling it out. Where a cheaper, smaller model does the job, use it for the bulk work and reserve the expensive one for the parts that need judgement. That kind of sizing is ordinary discipline in any automated workflow.

Do and do not

Do

  • Measure one real request before scaling a task up
  • Send the relevant extract, not the whole document
  • Start a fresh conversation for a new job

Do not

  • Budget by word count and expect it to match
  • Assume non-English text costs the same to process
  • Run everything through your most expensive model

Questions people ask about this

How do I know how many tokens my request will use?

Providers show usage in their dashboards after each request, and most publish a tokenizer page where you can paste text and see how it is split. For planning, run one realistic example, note what it consumed going in and coming back, then multiply by the volume you expect. Estimating from word counts alone will understate it, especially for non-English text.

Why does my AI assistant forget what I told it earlier?

Because a model can only hold a limited amount of a conversation at once, measured in tokens. Once a long chat passes that limit, the oldest material stops being visible to it, including instructions you gave at the start. Nothing warns you. Put standing rules somewhere they are repeated with each request, or start a new conversation and restate them.

Does writing in Nepali cost more than writing in English?

Usually yes, for processing. Text in scripts and languages that appear less often in training data tends to be broken into more tokens per word, so the same message consumes more of both your budget and the available context. If cost matters at volume, it is worth measuring both languages on your own content rather than assuming they are equivalent.

Related terms

Found this useful?

Share it, or ask an AI to summarise it

Back to the glossary

Knowing the term is the easy part

Applying it to your own site and budget is the work. Book a call and I will tell you what actually applies to you.