Polyaxon is an AI infra/orchestration platform for building, training, and monitoring large-scale ML applications. It supports Kubernetes-based deployment, multi-framework work with distributed training, and hyperparameter optimization features.
Collecting history — the radar snapshots this repo daily. The trend line appears after 3 days of data (1 so far).
What it is
Polyaxon is an AI Infra / AI Orchestration / AI Control Plane that deploys into data centers or cloud providers and supports major deep learning frameworks. It manages workloads with container and node management to utilize GPU servers as shared resources and provides a dashboard and experiment tracking capabilities. The repository has 3717 stars and 330 forks, with 126 open issues.
How it works
Polyaxon deploys into Kubernetes environments and provides tooling to run experiments, notebooks, and dashboards. It includes components for distributed training across TensorFlow, PyTorch, MPI, Horovod, and other frameworks, along with an automation/optimization engine for hyperparameter tuning (grid search, random search, Hyperband, Bayesian optimization, Hyperopt, iterative methods) and a DAG-based workflow engine for ML pipelines.
Getting started
Install the CLI and deploy a Polyaxon cluster as shown in the README:
# Install Polyaxon CLI
$ pip install -U polyaxon
# Create a namespace
$ kubectl create namespace polyaxon
# Add Polyaxon charts repo
$ helm repo add polyaxon https://charts.polyaxon.com
# Deploy Polyaxon
$ polyaxon admin deploy -f config.yaml
# Access API
$ polyaxon port-forward
# Create a project
$ polyaxon project create --name=quick-start --description='Polyaxon quick start.'
# Upload code and start experiments
$ polyaxon run -f experiment.yaml -u -l
# Start Polyaxon dashboard
$ polyaxon dashboard
# Start Jupyter notebook for your project
$ polyaxon run --hub notebook
# Start TensorBoard for a run's output
$ polyaxon run --hub tensorboard -P uuid=UUID
Please check the quick start guide linked in the README for more details.
Recent releases
- The latest releases section shows none explicitly in the provided data.
Traction
Stars: 3717 (1 day / 7 days data not provided in FACTS). Open issues: 126. Forks: 330.
Behind the repo
Not included in the provided facts.
Caveats
License: Apache-2.0. Created: 2016-12-26. Last push: 2026-08-03. Language: MDX. Topics include agents, AI/ML, Kubernetes, notebooks, and ML pipelines. No explicit licensing caveats beyond the Apache-2.0 license shown in the README. No age-related caveats beyond creation date.