Skip to content
AImpact
IT EN
Inference Beginner Also known as: Token

Token

The basic unit the model breaks text into: it can be a whole word, a syllable, or a few characters, depending on the tokenizer.

ShareLinkedInX

In practice

LLM APIs charge per input and output token. In English 1 token is roughly 0.75 words, in Italian a bit less. Counting tokens in your prompt helps estimate cost and stay within the context limit.

Related terms

Tokenizer Context window LLM

Seen in the wild

9 entries mentioning it
  1. SpaceXAI unveils Grok 4.5, a coding and agents model at 2/6 dollars per million tokens
    High
  2. Cerebras beats Nvidia Blackwell on Llama 4 Maverick: 2,522 tokens/s
    Medium
  3. NVIDIA at CoRL 2024: Isaac Lab generally available, six Project GR00T workflows and the Cosmos tokenizer
    Medium
  4. Emu3: a single transformer for text, images and video using only next-token prediction
    High
  5. Gemini 1.5 Pro: 1 million tokens in context
    High
  6. AudioPaLM: Google fuses PaLM-2 and AudioLM into a model that listens and speaks
    High
  7. Rebuff: four-layer open source prompt injection detector with canary tokens
    Medium
  8. EnCodec: Meta AI introduces a high-fidelity neural audio codec
    High
  9. HuggingFace Transformers 3.0: one API for slow and fast tokenizers
    Medium
← All terms