Prefect

Open-source Python orchestration engine that turns decorated functions into scheduled, retried and monitored workflows, self-hosted or on Prefect Cloud.

Visit prefect.io ↗Alternatives

Works with AI agents:llms.txt MCP

Ask your AI about this, with this page as the source:ChatGPT ↗Claude ↗Perplexity ↗

Prefect is for Python code that has to run on a schedule or on demand and recover from failures: data pipelines, report jobs, model training and other scripts that outgrew cron. You keep writing ordinary Python and add decorators, and Prefect tracks every run, retries failures and shows the state of everything in a UI.

A flow is a Python function decorated with @flow; inside it you call tasks, functions decorated with @task that can be cached, retried and run concurrently. Flows are not declared as a fixed DAG up front, so ordinary Python branching and loops decide what runs. A deployment attaches a schedule or trigger to a flow, and a work pool decides where its runs execute: on a worker you run in a process, Docker or Kubernetes, pushed straight to serverless platforms such as ECS, Cloud Run or Modal, or on Prefect-managed infrastructure.

What you get: retries with delays, task caching, a UI with run history and logs, automations, and background tasks that a web app can hand off with .delay() to a separate worker. The same Python client talks to either Prefect Cloud or a self-hosted Prefect server; the server stores state in SQLite by default and in PostgreSQL for production, with Redis added when you run several server instances.

The limits: workflows are Python only, and the install docs ask for Python 3.11 or newer. Prefect Cloud adds RBAC, audit logs and SSO that the open-source server does not have, and its plans cap users, deployments and how long run history is kept, 7 days on the free Hobby plan.

Where it fits

Background Jobs & Cron · Data pipeline orchestratorTurning existing Python scripts into scheduled, retried, monitored workflows with decorators, using ordinary Python control flow instead of a fixed DAG.Trade-off: Python only; RBAC, SSO and audit logs are Cloud-only, and the free Cloud plan caps users and deployments and keeps 7 days of run history.

Who uses it

1 makers' products, each linked to the source that shows it, and 1 open-source projects that declare it in their code.

The maker says so 1Declared in code 1How evidence is collected →

Sort

KypsoYour engineering workflows, reimagined

“We use Prefect to run our workflows and it's super flexible and easy to use!”

Background Jobs & CronMaker says so · source ↗

sparrow Structured data extraction, instruction calling and agentic workflows with ML, LLM and V…Background Jobs & CronIn its code · source ↗

Open source: a project that declares Prefect as a dependency in its public code — verifiable, but not necessarily a live product.

What makers say

1 makers on why they use Prefect, in their own words on Product Hunt.

We use Prefect to run our workflows and it's super flexible and easy to use!

Kypso Sep 2026 ↗

Alternatives to Prefect

All alternatives by situation →

Trigger.devLong-running TypeScript tasks — AI calls, media processing — that would time out in a serverless function.Open sourceFree tier23 makers · background jobs & cron

InngestEvent-driven, multi-step workflows where each step retries on its own and your code stays in your existing app.Open sourceFree tier16 makers · background jobs & cron

TemporalLong-running, multi-step workflows — payments, onboarding, AI agent loops — whose state must survive crashes and deploys, in Go, Java, Python, TypeScript and more.Open sourceFree tier35 makers · background jobs & cron

UpstashSimple scheduled calls and delayed HTTP jobs on serverless or edge hosting, via QStash.Free tier2 makers · background jobs & cron

BullMQNode apps on a long-running server that already have Redis — free and fully in your control.Open sourceFree tier1 makers · background jobs & cron

RabbitMQA dedicated message broker for queues and pub/sub between services, in any language, or as the broker behind Celery.Open sourceFree tier3 makers · background jobs & cron

DagsterData pipelines described by the tables, files and models they produce, with lineage between them, schedules, sensors and native dbt integration, in Python.Open sourceFree tier3 makers · background jobs & cron

Questions makers ask about Prefect

Do I have to use Prefect Cloud?

No. The Python SDK can talk to Prefect Cloud or to a Prefect server you host, which comes with a database and UI. Both share a common set of capabilities; Cloud adds organization features such as RBAC, audit logs and SSO. source ↗

What database does a self-hosted server need?

SQLite is the default and suits lightweight, single-server setups with essentially no configuration. PostgreSQL 14.9 or higher, with the pg_trgm extension, is recommended for production, high availability and multi-server deployments. source ↗

How much does Prefect Cloud cost?

Hobby is free with 2 users, 5 deployments and 500 serverless compute minutes a month. Starter is $100 a month, Team is $100 per user per month, and Enterprise is custom. source ↗

Where does my flow code run?

Wherever the deployment's work pool points. Hybrid pools use a worker you run in a process, Docker or Kubernetes; push pools submit runs to serverless platforms such as AWS ECS, Google Cloud Run, Azure Container Instances or Modal; managed pools run on Prefect's infrastructure. source ↗

Can it replace a job queue for my web app?

For Python apps, partly. Prefect has background tasks; a worker started with .serve() runs a task whenever your code calls .delay() on it, with the same arguments as the task itself. source ↗

How long is run history kept on Prefect Cloud?

Metadata is kept for 7 days on Hobby and Starter, 14 days on Team, 30 days on Pro and a custom period on Enterprise, counted from when a run finishes. source ↗

Is Prefect free?

Yes — there is a free tier a small product can run on; paid use starts at $100/mo. source ↗

Is Prefect open source or self-hostable?

Open source, and you can self-host it. source ↗

Can AI coding agents work with Prefect?

It serves an llms.txt docs index; it has an official MCP server (io.github.PrefectHQ/prefect-mcp-server).

Who uses Prefect?

1 makers' products we track, each with a source, and 1 open-source projects declare it in their code. source ↗