metaflow
Metaflow: More AI and ML, Less Engineering
Decision gist · record as of 2026-08-14
Yes. Metaflow is actively maintained, has low install friction, carries a permissive license, and is backed by proven production use at scale (Netflix runs 3000+ projects on it). It solves a real problem—bridging the gap between notebook prototyping and production ML systems—with a mature, well-documented API. Install if you're building ML workflows that need to scale beyond a single machine or move from experimentation to production.AI-flagged interpretation of the facts on this page — verify before relying
Before you install
- Low install friction with a pure-Python wheel distribution.
- Actively maintained with a recent release (3 days old) and strong community backing; the repository shows 10213 stars and continuous development.
License · maintenance · safety
Apache Software License (permissive) — Licensed under Apache Software License (permissive), which allows commercial use, modification, and distribution with minimal restrictions—suitable for both proprietary and open-source projects.
last release 2026-08-11 (3 days) · last repo commit 2026-08-13 · 10,213 stars
0 known vulnerabilities (OSV.dev, 2026-08-14) · 930,396 downloads/mo, #4,702 on PyPI
Alternatives
Verify before relying
pip install metaflow
from metaflow import FlowSpec, step
class MyFlow(FlowSpec):
@step
def start(self):
self.next(self.end)
@step
def end(self):
pass
if __name__ == '__main__':
MyFlow()- Whether the package requires AWS/cloud credentials or infrastructure setup to use core features beyond local prototyping
- Performance characteristics and overhead when running on modest hardware or in resource-constrained environments
- Compatibility details with specific notebook environments (Jupyter, JupyterLab, etc.) beyond the stated notebook support
What it is and what it does
Metaflow is a framework for managing the full lifecycle of AI and ML projects, from rapid local prototyping to production deployment. It provides a Pythonic API that unifies code, data, and compute management, allowing teams to write workflows once and scale them across local machines, cloud infrastructure, or production orchestrators without rewriting. The framework handles experiment tracking, versioning, dependency management, and both horizontal and vertical scaling with support for CPUs and GPUs.
Originally developed at Netflix and now supported by Outerbounds, Metaflow is designed for teams of any size working on classical statistics, deep learning, or foundation models. It reduces the engineering burden of moving from notebook prototypes to reliable production systems by providing built-in support for distributed computing, failure recovery, checkpointing, and reactive orchestration. The package depends on requests and boto3, making it suitable for cloud-native workflows, particularly on AWS.
Use it for
- Prototype ML models locally in notebooks, then scale the same code to run on distributed cloud compute without modification
- Track experiments, versions, and artifacts across team members with built-in versioning and visualization
- Deploy data pipelines and ML workflows to production orchestrators with one-click deployment and event-driven triggering
- Run embarrassingly parallel or gang-scheduled compute jobs reliably across CPUs and GPUs with automatic failure recovery
- Manage dependencies and data access for large-scale data processing jobs handling petabytes of data
Worth the install?
AI-flagged interpretation of the facts on this page. Verify before relying on it.
Yes.
Metaflow is actively maintained, has low install friction, carries a permissive license, and is backed by proven production use at scale (Netflix runs 3000+ projects on it). It solves a real problem—bridging the gap between notebook prototyping and production ML systems—with a mature, well-documented API. Install if you're building ML workflows that need to scale beyond a single machine or move from experimentation to production.
Install
metaflow on PyPI
Before you install
Low install friction with a pure-Python wheel distribution. Actively maintained with a recent release (3 days old) and strong community backing; the repository shows 10213 stars and continuous development.
License in practice
Licensed under Apache Software License (permissive), which allows commercial use, modification, and distribution with minimal restrictions—suitable for both proprietary and open-source projects.
Quickstart
pip install metaflow
from metaflow import FlowSpec, step
class MyFlow(FlowSpec):
@step
def start(self):
self.next(self.end)
@step
def end(self):
pass
if __name__ == '__main__':
MyFlow()Verify before relying
- Whether the package requires AWS/cloud credentials or infrastructure setup to use core features beyond local prototyping
- Performance characteristics and overhead when running on modest hardware or in resource-constrained environments
- Compatibility details with specific notebook environments (Jupyter, JupyterLab, etc.) beyond the stated notebook support
Package facts
| License | Apache Software License permissive |
| Python support | Not specified |
| Install friction | Low. Pure-Python wheel |
| Runtime dependencies | 2 packagesrequestsboto3 |
| Maintenance | Actively maintained 3 days since the last release |
| Last repo commit | |
| First released | |
| Downloads | 930,396 / month, #4,702 on PyPI 30-day window, as of 2026-08-14 |
| Known vulnerabilities | None known OSV.dev, checked 2026-08-14 |
| Classifiers | Development Status :: 5 - Production/StableLicense :: OSI Approved :: Apache Software LicenseOperating System :: MacOS :: MacOS XOperating System :: POSIX :: LinuxProgramming Language :: Python :: 3Programming Language :: Python :: 3.10Programming Language :: Python :: 3.11Programming Language :: Python :: 3.12Programming Language :: Python :: 3.13Programming Language :: Python :: 3.6Programming Language :: Python :: 3.7Programming Language :: Python :: 3.8Programming Language :: Python :: 3.9 |
Evidence: metaflow-2.19.37-py2.py3-none-any.whl
Tags
Let your AI agent find packages like this
Example. Real query, live index.
You found this page by searching. An agent finds it by wishing: SkillFed indexes 14,416 PyPI packages by what they can do, searchable in plain language.
wish › “distributed computing for ml”
- metaflowMetaflow is a framework for building, prototyping, and deploying AI…
- flyteFlyte is a Python SDK for defining, orchestrating, and executing ML…
- dagster-mlflowIntegrates MLflow experiment tracking and model management with…
Give your agent the search over MCP, or paste the wish link into any chat.
More Distributed Computing packages
gRPC Python is an HTTP/2-based RPC framework that enables you to define and call remote procedures across network boundaries using protocol buffers for serialization.
Install it if you need RPC communication in a distributed system or are integrating with existing gRPC services.
execnet lets you spawn and communicate with Python interpreters across local processes, remote hosts, and different platforms, using a simple API for task distribution and inter-process messaging.
However, the aging maintenance status (275 days since last release) means you should verify it meets your concurrency and performance needs before committing to a…
Cloudpickle extends Python's standard pickle module to serialize lambda functions, interactively-defined functions and classes, and other constructs that the default pickle cannot handle, making it suitable for cluster computing and remote code execution.
Install it if you need to serialize lambda functions, interactively-defined code, or non-standard Python constructs for cluster computing or distributed execution.
Provides a unified, open()-compatible Python API for streaming large files from remote storage (S3, GCS, Azure, HDFS, SFTP, HTTP) and local filesystems, with transparent compression support.
Install it if you work with large files on cloud storage or remote systems and want to avoid writing boilerplate around multiple SDKs.
Portalocker provides cross-platform file locking with support for exclusive and shared locks, plus Redis-based distributed locks and process-aware PID file locking.
Install it if you need file or process coordination; the optional extras (pywin32, redis) are only required for specific lock types.
Ray is a distributed computing framework that scales Python applications from a single machine to multi-node clusters, providing abstractions for parallel tasks, stateful actors, and shared objects.
See also metaflow-stubs · ob-metaflow · metaflow-netflixext · ob-metaflow-extensions · outerbounds · metaflow-prebuilt · mlflow · flyte · azureml-pipeline · metaflow-torchrun