Concept
aka context length
aka context size
Context Window
Maximum number of tokens (prompt + completion) a model can attend to in a single request.
Definition
The context window is the hard upper bound on how much text — measured in tokens — fits into one model call. Modern LLMs range from 8k to 2M tokens. A larger context lets the model reason over longer documents, but compute cost scales super-linearly with context length, which is why long-context inference is expensive.
Common use cases
- Long document QA
- Codebase analysis
- Multi-turn chat
Related models
claude
gemini
gpt