loading
loading
The basic unit LLMs use to process text — roughly 3/4 of a word on average.
Models don't read characters or words; they read tokens, which are sub-word chunks learned during training. Costs, speed, and context window limits are all measured in tokens. Knowing your token usage matters when you're optimizing cost or working near a model's context limit.
Plainly
Think of Token as the brain part that guesses or decides. The basic unit LLMs use to process text — roughly 3/4 of a word on average.
In practice
Use it when model choice, prompts, latency, cost, or quality affect the product result. In practice, define the owner, input, output, and failure mode before you rely on it.