Token

A token is the smallest unit of text that a large language model processes. One token roughly corresponds to ~0.75 English words or 1-2 Chinese characters. Models don't see characters — they see token IDs.

Why it matters

• API billing: nearly all LLM services charge per token (input + output billed separately).
• Context window: a model's context window is measured in tokens (4k / 32k / 200k / 1M).
• Inference speed: tokens/s is the core throughput metric.

Rough estimates

Note: Chinese tokenization is typically 2-3x less efficient than English — one Chinese character may split into 1-3 tokens depending on the tokenizer.