HomeGlossary › Token Limit / Context Window

Token Limit / Context Window

A definition from the GenAI for Business glossary.

Every model has a maximum "context window", the total number of tokens (words and symbols) it can process at once. For example, current GPT flagships handle around 400k tokens and current Claude models up to 1M tokens (roughly a 1,500-page book). Longer context windows allow richer reasoning, but also cost more and take longer to compute.

Want the full picture? This term comes from GenAI for Business, a complete free book on generative AI strategy and implementation by Prof. Shubin Yu (HEC Paris).
Scaling Up vs. Scaling OutMultimodality