TypedDict
Python 标准库 typing 模块中的类型定义方式，用于描述具有固定键的字典结构，LangGraph 中常用其定义 Agent 的 State 对象
Timeline (last 90 days)
在 LangGraph 中，用 TypedDict 定义 State 时，messages 字段配合 add_messages reducer 可自动实现消息追加。
LangGraph uses Python's TypedDict to declare the graph's shared state structure, where each node function receives the complete state and returns a delta dictionary, with merge updates performed by field.
In Python 3.15, TypedDict supports declaring the type of extra items, and a new TypeForm aimed at library authors has been added.
The OpenAI SDK makes extensive use of TypedDict to describe request body structures.
NotRequired is a type annotation utility introduced in Python 3.11 and backward-compatible via typing_extensions, used specifically in TypedDict scenarios.
TypedDict allows developers to define precise key-value types for dictionaries; by default all fields are required, and NotRequired[T] can be used to mark specific fields as optional.
In LangGraph, State is typically defined using Python's TypedDict or Pydantic models, and is essentially a strongly-typed dictionary structure.
All Facts (7)
在 LangGraph 中，用 TypedDict 定义 State 时，messages 字段配合 add_messages reducer 可自动实现消息追加。
50%UnverifiedLangGraph uses Python's TypedDict to declare the graph's shared state structure, where each node function receives the complete state and returns a delta dictionary, with merge updates performed by field.
50%UnverifiedTypedDict allows developers to define precise key-value types for dictionaries; by default all fields are required, and NotRequired[T] can be used to mark specific fields as optional.
50%UnverifiedThe OpenAI SDK makes extensive use of TypedDict to describe request body structures.
50%UnverifiedNotRequired is a type annotation utility introduced in Python 3.11 and backward-compatible via typing_extensions, used specifically in TypedDict scenarios.
50%UnverifiedIn LangGraph, State is typically defined using Python's TypedDict or Pydantic models, and is essentially a strongly-typed dictionary structure.
50%UnverifiedIn Python 3.15, TypedDict supports declaring the type of extra items, and a new TypeForm aimed at library authors has been added.
50%