A Pythonic Alternative to Graph Construction

LangGraph v1.2+ introduced the Functional API, a radically simpler way to build agents using Python decorators instead of explicit graph construction. While the Graph API (StateGraph, nodes, edges) remains the gold standard for complex workflows, the Functional API lets you build production-grade agents with plain Python functions decorated with @entrypoint and @task.

Core Decorators

Functional API Example

from langgraph.func import entrypoint, task

@task
async def research(topic: str) -> str:
    """Search the web and compile research notes."""
    results = await web_search(topic)
    return summarize(results)

@task
async def write_draft(notes: str) -> str:
    """Write a blog post draft from research notes."""
    return await llm.generate(f"Write a blog post based on: {notes}")

@task
async def review(draft: str) -> dict:
    """Score the draft and provide feedback."""
    return await llm.generate_json(f"Score 1-10 and give feedback: {draft}")

@entrypoint
async def blog_writer(topic: str) -> str:
    notes = await research(topic)
    draft = await write_draft(notes)
    result = await review(draft)
    while result["score"] < 8:
        draft = await write_draft(notes + f"\nFeedback: {result['feedback']}")
        result = await review(draft)
    return draft

Functional API vs Graph API