metaflow-prebuilt
Metaflow extension: pre-bake conda environments into Docker images for fast cold starts
Decision gist · record as of 2026-08-14
Yes, if you run Metaflow flows on remote infrastructure (Batch, Kubernetes) and want to reduce task startup latency by pre-baking dependencies. The extension is actively maintained, has low install friction, and offers a clean configuration model. Install only if you have a registry and build service already available or planned; it adds no value for local-only flows.AI-flagged interpretation of the facts on this page — verify before relying
Before you install
- Requires Python 3.10 or later; requires a configured Docker registry and build service to be set up before deployment.
- Low install friction; wheel-based distribution.
- Active maintenance with a recent release 46 days ago.
License · maintenance · safety
Apache Software License (permissive) — Apache Software License is permissive, allowing commercial and private use with minimal restrictions; suitable for most production deployments.
last release 2026-06-29 (46 days) · last repo commit 2026-08-05 · 60 stars
0 known vulnerabilities (OSV.dev, 2026-08-14) · 98,470 downloads/mo, #13,082 on PyPI
Alternatives
Verify before relying
pip install metaflow-prebuilt
from metaflow import FlowSpec, step, conda
class MyFlow(FlowSpec):
@conda(packages={"numpy": "1.26.4"})
@step
def start(self):
self.next(self.end)
@step
def end(self):
pass
if __name__ == "__main__":
MyFlow()
# Deploy: python my_flow.py --environment=prebuilt batch run- Whether optional extras (ecr, gcr, kaniko) are required for specific cloud backends or if base install works with environment variables alone.
- Performance improvement magnitude compared to standard conda bootstrap in typical workflows.
- Compatibility with metaflow versions prior to or after the first_release date of this extension.
What it is and what it does
metaflow-prebuilt is a Metaflow extension that shifts dependency resolution and Docker image building from task runtime to deployment time. Instead of each task bootstrapping its conda environment when it starts, the extension resolves the environment spec for each step during deployment, builds a Docker image with all dependencies pre-installed, pushes it to a registry, and configures the remote runner to pull that image. This eliminates the conda bootstrap overhead at task startup, reducing cold-start latency.
The extension supports multiple build services (local Docker, Buildx, Kaniko, AWS CodeBuild) and image registries (Docker Hub, ECR, GCR, local), configured via environment variables. It is designed for teams running Metaflow flows on Batch, Kubernetes, or other remote runners where pre-built images provide measurable startup improvements. The extension is extensible: you can subclass DockerBuildService or ImageRegistry to add custom backends.
Use it for
- Deploy Metaflow flows on AWS Batch with pre-built ECR images to reduce task cold-start time.
- Run Metaflow workflows on Kubernetes using Kaniko to build and cache images without a local Docker daemon.
- Integrate Metaflow with AWS CodeBuild for centralized image builds and push to ECR in CI/CD pipelines.
- Develop and test Metaflow flows locally using a local registry container before pushing to production.
- Extend the build or registry layer with custom implementations for proprietary image storage or build infrastructure.
Worth the install?
AI-flagged interpretation of the facts on this page. Verify before relying on it.
Yes, if you run Metaflow flows on remote infrastructure (Batch, Kubernetes) and want to reduce task startup latency by pre-baking dependencies.
The extension is actively maintained, has low install friction, and offers a clean configuration model. Install only if you have a registry and build service already available or planned; it adds no value for local-only flows.
Install
metaflow-prebuilt on PyPI
Before you install
Low install friction; wheel-based distribution. Active maintenance with a recent release 46 days ago. Depends on metaflow and metaflow-netflixext, both of which must be available in your environment.
Requires Python 3.10 or later; requires a configured Docker registry and build service to be set up before deployment.
License in practice
Apache Software License is permissive, allowing commercial and private use with minimal restrictions; suitable for most production deployments.
Quickstart
pip install metaflow-prebuilt
from metaflow import FlowSpec, step, conda
class MyFlow(FlowSpec):
@conda(packages={"numpy": "1.26.4"})
@step
def start(self):
self.next(self.end)
@step
def end(self):
pass
if __name__ == "__main__":
MyFlow()
# Deploy: python my_flow.py --environment=prebuilt batch runVerify before relying
- Whether optional extras (ecr, gcr, kaniko) are required for specific cloud backends or if base install works with environment variables alone.
- Performance improvement magnitude compared to standard conda bootstrap in typical workflows.
- Compatibility with metaflow versions prior to or after the first_release date of this extension.
Package facts
| License | Apache Software License permissive |
| Python support | Supports the current Python release >=3.10 |
| Install friction | Low. Pure-Python wheel |
| Runtime dependencies | 2 packagesmetaflowmetaflow-netflixext |
| Maintenance | Actively maintained 46 days since the last release |
| Last repo commit | |
| First released | |
| Downloads | 98,470 / month, #13,082 on PyPI 30-day window, as of 2026-08-14 |
| Known vulnerabilities | None known OSV.dev, checked 2026-08-14 |
| Classifiers | Development Status :: 4 - BetaLicense :: 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 :: 3.13 |
Evidence: metaflow_prebuilt-0.3.5-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 › “metaflow docker environment caching”
- metaflow-prebuiltMetaflow extension that pre-bakes conda and PyPI dependencies into…
- metaflow-netflixextProvides Netflix-developed extensions to Metaflow, including an…
- metaflow-checkpointProvides decorators for Metaflow workflows to save and resume…
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-torchrun · metaflow-netflixext · ob-metaflow-extensions · metaflow · metaflow-stubs · metaflow-checkpoint · ob-metaflow · ob-metaflow-stubs · kfp-pipeline-spec · argo-workflows