Context window

As explained in AI Agents in Action

A context window is the maximum amount of tokenized input and related context that a language model can consider in one interaction.

A context window is the maximum amount of tokenized input and related context a language model can consider in one interaction. It determines how much context or memory the model can hold. section 2.4

Choose context capacity according to the task:

• A window below 4,000 tokens is typically enough for simple tasks. section 2.4
• A larger window can become essential when multiple agents converse about one task. section 2.4
• Context-window sizes vary across model architectures and model types. section 2.4

Selecting a model for an agent

For a simple task, a model with a smaller context window may be sufficient. For an agent system in which several agents exchange messages, prioritize a model offering a larger window. section 2.4

Figure 2.11 The important criteria to consider when consuming an LLM

Compare context size with other criteria when selecting an LLM for an agent. section 2.4

How AI Agents in Action covers it

How deep it goes

Defines context capacity and gives practical thresholds, but does not explain tokenization or model internals.

Examples

Uses simple tasks below 4,000 tokens and conversations among multiple agents as contrasting capacity requirements.

What you can do after

Evaluate model options by estimating whether their context capacity fits simple tasks or multi-agent conversations.

How it connects

Connects context windows with model architecture, memory, and multi-agent systems.

liveBookmanning.com

Read AI Agents in Action on liveBook Get the book at manning.com

Also explained in

• 

  liveBookmanning.com
  Software Testing with Generative AIThe bounded amount of tokenized input and related context that a language model can process for a request.
• 

  liveBookmanning.com
  Generative AI in ActionThe context window is the bounded amount of input and generated text that a language model can consider within one interaction.
• 

  liveBookmanning.com
  LLMs in ProductionContext length is the maximum amount of tokenized input and generated context that a language model can process in one request.
• 

  liveBookmanning.com
  AI for Everyday ITA context window is the amount of input and conversational or document context that a language model can process for one interaction.
• 

  liveBookmanning.com
  Time Series Forecasting Using Foundation ModelsContext length is the fixed number of preceding sequence elements that a model can consider when processing an input.
• 

  liveBookmanning.com
  Starting Data Analytics with Generative AI and PythonThe maximum amount of conversational or input-output context that a language model implementation can process in one exchange.
• 

  liveBookmanning.com
  AI Agents in Action, Second EditionThe finite amount of input and conversational context that an agent can retain or process in one interaction.
• 

  liveBookmanning.com
  Grokking AI Algorithms, Second EditionA context window is the maximum sequence of tokens a language model processes as contextual input at one time.
• 

  liveBookmanning.com
  Build AI-Enhanced Web AppsA context window is the maximum amount of tokenized input and generated content a language model can process in one interaction.
• 

  liveBookmanning.com
  Grokking AI ApplicationsA context window is the maximum number of tokens a language model can process in one prompt.
• 

  liveBookmanning.com
  The Art of AI Product DevelopmentThe maximum amount of tokenized input and retained conversational or task information that a language model can process in one operation.
• 

  liveBookmanning.com
  Machine Learning Platform EngineeringA context window is the maximum amount of tokenized input and output that a language model can process in one interaction.
• 

  liveBookmanning.com
  AI Agents and ApplicationsA context window is the maximum amount of input and conversational context a language model can process in one interaction.
• 

  liveBookmanning.com
  Effective Conversational AIA context window is the maximum amount of input text or tokens that a language model can process in one inference request.
• 

  liveBookmanning.com
  Build Python Web Apps with StreamlitContext window length is the maximum number of tokens that a language model can process as input in a single prompt.
• 

  liveBookmanning.com
  Building Reliable AI SystemsThe maximum amount of input content a model can process in a single input.
• 

  liveBookmanning.com
  Build Applications with Local AI Models on a MacContext length is the maximum number of tokens a language model can consider at once.
• 

  liveBookmanning.com
  How Large Language Models WorkContext size is the maximum number of tokens a language model can computationally process in one request.
• 

  liveBookmanning.com
  Build an AI Agent (From Scratch)A context window is the maximum amount of tokenized input and generated context that a language model can process together.
• 

  liveBookmanning.com
  Spring AI in ActionA context window is the maximum amount of tokenized input and output that a language model can handle in one interaction.
• 

  liveBookmanning.com
  Build a Large Language Model (From Scratch)The number of preceding tokens the model conditions on when predicting the next token.

Compare how 22 books explain context window →

Related concepts

• tokenmeasures · A token is a discrete unit of text, such as a word, word fragment, character, or symbol, represented and processed as an element by a language model.
• large language modelused by · A large language model is a machine-learning model, typically neural, trained on extensive text data to process and generate language.
• contextcontrasts with · Context is the surrounding information, state, or circumstances that condition how an entity, event, or task is interpreted or how a system behaves.

This explainer was generated by Manning's knowledge graph from the text of AI Agents in Action. The section links point to the parts of the book it draws on.