Term
Context window
The context window is the maximum number of tokens a language model can process at once — input and output combined.
Context window — explained in more detail
A language model’s context window defines how much text it can keep in view within a single request. The input prompt and the model’s response share this budget. Once it fills up, older content is either truncated or has to be supplemented through external mechanisms such as retrieval.
Practical relevance
Sizes vary widely. Smaller local models often sit between 4,000 and 32,000 tokens. GPT-4o is around 128,000 tokens, while current Claude models reach up to one million. Anyone wanting to process large codebases or long documents in a single request needs to size accordingly.
Distinction from similar terms
The context window is not the model’s training-data volume — the latter is orders of magnitude larger and fixed. It is also distinct from hardware RAM: the context window is a model property; memory consumption scales with it but is a separate concern of the inference infrastructure.
Entdecke mehr
AI Workflows by Keyword: How We Make Recurring Routines Enforceable
A typed keyword triggers a fixed AI routine — and every single step must be committed before the next one appears. Why that's the actual trick.
GlossarReasoning Effort
Reasoning Effort is a control parameter on modern reasoning models that sets how much internal step-by-step thinking (thinking tokens) a model spends before answering — higher levels raise quality but also latency and cost.
LexikonLLM Hallucinations — Causes and Remedies
Why LLMs confidently invent falsehoods, what types of hallucinations exist, and which remedies actually help — RAG, source enforcement, verification.