Added Nov 5, 2026
Polyaxon
Freemium · $555/moOpen-source control plane for ML training, serving and agents on Kubernetes
What is Polyaxon?
Polyaxon schedules, tracks and governs machine learning work on your own Kubernetes clusters, from GPU training and fine-tuning to model serving and agent sandboxes. It is an open-source platform with paid tiers for teams that need more seats and controls.
Jobs are described in Polyaxonfiles written in YAML, JSON, Python, Go or TypeScript, and cover fine-tuning with LoRA, RL, hyperparameter sweeps, DAGs and batch runs from one GPU to many nodes. Services such as inference endpoints, notebooks and TensorBoard share the same queues and data connections.
Teams get experiment tracking, comparison dashboards, evals and a model registry, plus audit logs and access policies. Because it runs on your own clouds, on-prem clusters and storage, training data stays in infrastructure you control.
Features
- Training and fine-tuning jobs
- Hyperparameter sweeps and DAGs
- Model serving and notebooks
- Experiment tracking and registry
- Agent sandboxes
- Access policies and audit logs
Tags
Releases
+ Submit new release- Listed
Polyaxon was added to the directory.
Organization
Polyaxon
Listed by kitio.ai · unverified owner
Pricing
- Pricing model
- Freemium
- Paid options from
- $555/mo
- Billing frequency
- Monthly
Polyaxon plans
Open Source
Free
- Self-hosted
- Core tracking and scheduling
Platform
$555/mo
- 3 developer seats
- 1 compute cluster
- 50 concurrent runs
- Collaboration features
Teams
$1500/mo
- 6 developer seats
- 2 compute clusters
- Teams management
- Priority support
Enterprise
Custom
- Custom seats
- Custom policies and deployment
Prices in USD, checked Nov 5, 2026. Plans change often, so confirm on the Polyaxon site.Official pricing
Sponsored
Your AI tool here
A Featured launch puts your tool in this spot and on the home page for a week.
Top Polyaxon alternatives
View allModels & APIs
Open-source platform for ML experiments, models and GenAI tracing
Models & APIs
Experiment tracking and LLM observability, now part of CoreWeave
Models & APIs
Open-source Python framework for ML and AI workflows
Compare Polyaxon
If you liked Polyaxon
- ClearMLNew
Open-source MLOps and AI infrastructure platform for GPU clusters
Freemium · $15/moModels & APIs - NannyMLNew
Monitors deployed ML models and estimates performance without labels
FreemiumModels & APIsMade in Belgium - MLServerNew
Open-source Python server for serving ML models over REST and gRPC
FreeModels & APIsMade in United Kingdom Open-source model serving and a managed platform for AI inference
FreemiumModels & APIsMade in United States
Pros and cons
Pros
- Self-hosted, data stays on your infra
- Covers training, serving and tracking
- Free for academics
Cons
- Paid plans start at a high price
- Requires Kubernetes expertise
- Extra seats and clusters cost more
Prompts & results
No prompts yet. Add your own prompt and output to help others use this tool.
Share a prompt
Reviews
Q&A
What is Polyaxon?
Polyaxon schedules, tracks and governs machine learning work on your own Kubernetes clusters, from GPU training and fine-tuning to model serving and agent sandboxes. It is an open-source platform with paid tiers for teams that need more seats and controls.
Can I self-host Polyaxon for free?
Yes. The core platform is open source and self-hostable, and academics can get the paid platform free by contacting support.
What file formats does Polyaxon use to define jobs?
Runs are described in Polyaxonfiles, which can be written in YAML, JSON, Python, Go or TypeScript.
How much does Polyaxon cost?
Polyaxon is freemium, with paid plans from $555/mo.
Is there a free version of Polyaxon?
Yes. Polyaxon has a free plan, and paid plans add more usage and features.
What are the main features of Polyaxon?
Training and fine-tuning jobs, Hyperparameter sweeps and DAGs, Model serving and notebooks, Experiment tracking and registry, Agent sandboxes, Access policies and audit logs.
Who makes Polyaxon?
Polyaxon is made by Polyaxon.
What are the best alternatives to Polyaxon?
Popular alternatives include MLflow, Kubeflow, Weights & Biases.