详细内容
A framework for real-life ML, AI, and data science
Open-source Metaflow makes it quick and easy to build and manage real-life
ML, AI, and data science projects.


Modeling
Use any Python libraries for models and business logic. Metaflow
helps manage libraries locally and in the cloud.
Deployment
Deploy workflows to production with a single command and
integrate with surrounding systems seamlessly.
Versioning
Metaflow tracks and stores variables inside the flow
automatically for easy experiment tracking and debugging.
Orchestration
Create robust workflows in plain Python. Develop and debug
them locally, deploy to production without changes.
Compute
Leverage the cloud to execute functions at scale. Use GPUs,
multiple cores, and large amounts of memory as needed.
Data
Access data from data warehouses. Metaflow flows
data across steps, versioning everything on the way.
Metaflow is used by these companies and hundreds of others
Metaflow is built for ML/AI engineers and data scientists, not just for machines
Develop with Metaflow
Explore with notebooks, develop with Metaflow, and test and debug
locally. Results are stored and tracked automatically for easy
analysis.

-
Scale out to the cloud
Break out from the confines of a laptop or a single notebook. Scale out
easily to the cloud, utilizing GPUs, multiple cores, and multiple instances
in parallel. Metaflow organizes the work for easy collaboration on the way.

Deploy to production confidently
Deploy experiments to
production with a single click without changing anything in the code.
Make flows react to updating data and other events automatically.
Bring your own Cloud
Get started easily on a laptop. When you are ready to scale, deploy the Metaflow stack
on your cloud account or on-premise Kubernetes cluster. Metaflow integrates
seamlessly with your existing infrastructure, security, and data governance
policies.
To get a taste of Metaflow in the cloud, try
Metaflow Sandbox
in the browser.
\
\
AWS \
\
Deploy on EKS and S3, or AWS Batch & AWS Step Functions.
\
\
Azure \
\
Deploy on AKS and Azure Blob Storage.
\
\
Google Cloud \
\
Deploy on GKE and Google Cloud Storage.
\
\
Kubernetes \
\
For maximum flexibility, deploy on a custom Kubernetes cluster.
Battle-hardened at Netflix
Metaflow was originally developed at Netflix to address the needs
of developers and data scientists who work on demanding real-life ML,
AI, and data projects. Netflix open-sourced Metaflow in 2019.
Today, Metaflow is used by hundreds of companies across industries,
powering diverse projects from state-of-the-art GenAI and compute vision
to business-oriented data science, statistics, and operations research.
- Open-Sourcing Metaflow, a Human-Centric Framework for Data\
Science - Unbundling Data Science Workflows with Metaflow and AWS Step\
Functions - Open-Sourcing a Monitoring GUI for Metaflow, Netflix’s ML\
Platform - Supporting content decision makers with machine learning
How leading ML, AI, and data science teams use Metaflow
\
\
Developing safe and reliable ML products at 23andMe \
\
Our complex, multi-stage workflows are codified and orchestrated using Metaflow.\
\
Accelerating ML within CNN\
\
Our data science team believes they were able to test twice as many models in Q1 2021 as they did in all of 2020.\
\
Accelerating experimentation with MLOps\
\
Metaflow helped us avoid the anti-pattern of needing to push code to find out if something works.\
\
Improving Data Science Processes to Speed Innovation at Realtor.com\
\
The team has shaved months off the time it takes to build a productionized machine learning model.
Recent release highlights
- Develop flows quickly with spin\
\
Create flows incrementally step-by-step with the new spin command\
\
November 4th, 2025 - Support for recursive and conditional steps\
\
Build agentic systems with the new recursive and conditional steps\
\
August 27th, 2025 - Develop custom decorators\
\
Compose flows with reusable custom decorators\
\
July 14th, 2025 - Support for uv\
\
Use uv to manage dependencies, from dev to cloud\
\
May 21st, 2025 - One-click local development stack\
\
Setup the full Metaflow stack on your laptop with one click\
\
March 5th, 2025 - Checkpointing progress\
\
Checkpoint long-running model training and other tasks with the new @checkpoint decorator\
\
February 9th, 2025 - Configurable Metaflow\
\
Configure flows freely with the new Config object\
\
December 19th, 2024 - Run and deploy flows programmatically\
\
New APIs allow you to run and deploy Metaflow in notebooks and scripts\
\
July 25th, 2024 - New Documentation for Compute Patterns\
\
Learn about various patterns of scalable compute with Metaflow.\
\
May 4th, 2024 - Support for AWS Trainium\
\
Train and fine-tune large language models and other\
generative AI models on AWS Trainium.\
\
April 29th, 2024 - Real-Time, Dynamic Cards\
\
Build observable ML/AI systems with cards that update in real-time.\
\
January 19th, 2024




