Skip to content

Examples

Learn by example with these practical use cases.

Overview

Example Description
Basic Usage Getting started with both processors
Advanced Custom counters, prompts, and configurations
Context Manager Real-time token tracking and tool truncation

Quick Examples

Minimal Summarization

Python
from pydantic_ai import Agent
from pydantic_ai_summarization import create_summarization_processor

processor = create_summarization_processor()

agent = Agent(
    "openai:gpt-4.1",
    history_processors=[processor],
)

Minimal Sliding Window

Python
from pydantic_ai import Agent
from pydantic_ai_summarization import create_sliding_window_processor

processor = create_sliding_window_processor()

agent = Agent(
    "openai:gpt-4.1",
    history_processors=[processor],
)

With Custom Trigger

Python
processor = create_summarization_processor(
    trigger=("messages", 30),
    keep=("messages", 10),
)
Python
processor = create_sliding_window_processor(
    trigger=("messages", 60),
    keep=("messages", 30),
)

With Multiple Triggers

Python
from pydantic_ai_summarization import SummarizationProcessor

processor = SummarizationProcessor(
    model="openai:gpt-4.1",
    trigger=[
        ("messages", 50),
        ("tokens", 100000),
    ],
    keep=("messages", 20),
)
Python
from pydantic_ai_summarization import SlidingWindowProcessor

processor = SlidingWindowProcessor(
    trigger=[
        ("messages", 100),
        ("tokens", 50000),
    ],
    keep=("messages", 30),
)

Choosing a Processor

Scenario Recommended
Context quality matters SummarizationProcessor
Speed/cost matters SlidingWindowProcessor
Many parallel conversations SlidingWindowProcessor
Coding assistant SummarizationProcessor
Simple chatbot SlidingWindowProcessor

Next Steps