temporalio

What it is and what it does

Temporal Python SDK is a framework for authoring workflows and activities that run on the Temporal orchestration engine. It turns async Python functions into durable, distributed workflows backed by a fault-tolerant event loop, allowing long-running business logic to survive failures through automatic replay and retry. The SDK integrates with asyncio, supporting async, threaded, and multiprocess activity execution patterns.

Workflows are defined as async functions decorated with @workflow.defn, and activities as functions decorated with @activity.defn. The framework handles task management, sleep, cancellation, and state persistence transparently. It provides type-safe workflow invocation through generics and MyPy support, custom data conversion (including Pydantic support), signal and update handlers, child workflows, timers, and built-in testing utilities with automatic time skipping.

Use it for

• Orchestrating multi-step business processes that span hours or days with automatic failure recovery and retry logic.
• Building resilient microservice workflows where individual steps may fail but the overall process completes reliably.
• Scheduling and executing long-running data processing or ETL jobs with built-in checkpointing and resumption.
• Implementing saga patterns for distributed transactions across multiple services with compensation on failure.
• Creating approval workflows or human-in-the-loop processes that pause and resume based on external signals.

Worth the install?

AI-flagged interpretation of the facts on this page. Verify before relying on it.

Install

temporalio on PyPI

Before you install

Medium install friction due to compiled wheels for multiple platforms. Active maintenance with recent releases; last commit 2026-08-13 and 1162 repository stars indicate ongoing development and community use.

Requires Python 3.10 or later; a running Temporal server instance is needed to execute workflows.

License in practice

MIT license permits commercial and private use with minimal restrictions, making it suitable for most production deployments.

Quickstart

pip install temporalio

from temporalio import activity, workflow
from datetime import timedelta

@activity.defn
def say_hello(name: str) -> str:
    return f"Hello, {name}!"

@workflow.defn
class SayHello:
    @workflow.run
    async def run(self, name: str) -> str:
        return await workflow.execute_activity(
            say_hello, name, schedule_to_close_timeout=timedelta(seconds=5)
        )

Package facts

Evidence: temporalio-1.31.0-cp310-abi3-macosx_10_12_x86_64.whl; temporalio-1.31.0-cp310-abi3-macosx_11_0_arm64.whl; temporalio-1.31.0-cp310-abi3-manylinux_2_17_aarch64.manylinux2014_aarch64.whl; temporalio-1.31.0-cp310-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl; temporalio-1.31.0-cp310-abi3-musllinux_1_2_aarch64.whl; temporalio-1.31.0-cp310-abi3-musllinux_1_2_x86_64.whl; temporalio-1.31.0-cp310-abi3-win_amd64.whl

Tags

More Distributed Computing packages

See also dbos · simpleflow · conductor-python · gen-worker · mistralai-workflows · agent-framework-durabletask · durabletask · azure-functions-durable · durabletask.azuremanaged · langgraph