Context Window

Definitions

1. The maximum number of tokens (input + output) that a language model can process in a single interaction. Determines how much text the model can 'see' at once. Larger context windows enable processing entire codebases or documents but increase compute cost quadratically.

   In plain English: The maximum amount of text an AI can read and consider at once — like the size of its working memory.

   Example: "Claude's 200K context window means you can paste an entire codebase and ask it to find bugs across files."

Related Terms

• Context Window
• mixture of agents
• system prompt
• Temperature
• Agentic Workflow
• Guardrails
• Retrieval-Augmented Generation
• LoRA
• Attention
• Prompt Injection
• Tool Use
• Multimodal AI
• AI Agent
• Adapter Layer
• Chain of Thought Prompting
• Differential Transformer
• DPO
• Dual Encoder
• Function Calling
• Gemma
• GPT Architecture
• Graph Transformer
• Hypernetwork
• In-Context Learning
• Large Multimodal Model
• PII Detection
• Responsible AI
• Safety Classifier
• Safety Filter
• Super Alignment
• Value Alignment
• Gating Network
• Neural Network Depth