Union.ai

What is Union.ai?

Union.ai is the enterprise offering built around Flyte, an open-source runtime for durable AI and data workflows. Union hosts the control plane, while your workloads, data, logs and code stay in your own cloud account on AWS, GCP or Azure. Teams use it across the ML lifecycle, from data preparation through model training and evaluation to model serving, with caching, retries and recovery so interrupted runs do not repeat finished work. You bring Flyte code as it is, and Union adds features such as Zero Trust security, remote debugging, replay and recover, federated login with SCIM, plus published usage pricing. The Team plan has a $950 monthly minimum that is credited back as usage, waived in the first month, and usage is metered on actions, vCPU-hours, memory and GPU-hours. Enterprise is custom and adds multi-cloud, SSO and SAML and a self-hosted option, while Flyte itself is free under Apache 2.0.

Use cases

• Orchestrate ML training pipelines
• Run AI agents and LLM endpoints
• Evaluate models before promotion
• Serve batch and real-time inference
• Keep AI workloads in your own cloud
• Scale Flyte beyond one cluster

Who it's for

• ML engineers
• Platform teams
• Data engineering teams
• Regulated enterprises

Why use Union.ai

• Data, logs and code stay in your own storage while Union runs the control plane
• Open-source Flyte code runs unchanged, so there is a path from self-run to managed
• Usage prices are published, with bands that fall as volume grows
• Your first month of usage is covered up to $950
• Workloads recover from interruptions without redoing completed work

Union.ai features

• Managed control plane with data plane in your cloud
• Durable tasks with caching, retries and recovery
• Remote debugging, replay and recover
• Versioned datasets, checkpoints and evaluation reports
• Zero Trust security and observability
• Federated login with SCIM
• Reusable containers and Ray clusters
• Published pricing per action, vCPU, memory and GPU
• Union-managed clusters at $500 per cluster per month, optional
• SSO, SAML and custom RBAC on Enterprise

Union.ai integrations

• AWS
• Google Cloud
• Microsoft Azure
• Kubernetes
• Ray
• AWS Marketplace
• Flyte

How to get started with Union.ai

1. Subscribe through AWS Marketplace, or contact Union for GCP or Azure.
2. Sign up with a Google or Microsoft work account and create your organization.
3. Run a first workflow locally with the Flyte CLI and follow it in the Union UI.
4. Provision a Kubernetes cluster, storage and identities in your cloud account.
5. Connect the cluster so Union installs the data plane, then run workloads on it.

Union.ai limitations

• Team plan caps: 2 active clusters, 10,000 actions running at once, and up to 200,000 actions in each run
• Team has 30-day data retention and 12x5 standard support
• Direct sign-up was planned to follow the AWS Marketplace launch
• Reserved or owned GPUs require talking to sales

Union.ai pricing

Union.ai has no free plan, but it offers a free trial. Paid plans start at $950 per month, according to the official pricing page on 24 Oct 2026.

Union.ai plans

• Flyte

  Free

  open source, Apache 2.0

  • Full Flyte SDK and engine
  • Flyte UI
  • Community support

• Team

  $950/month minimum

  credited to usage; waived in first month

  • 2 active clusters
  • 10,000 concurrent actions
  • 200,000 actions per run
  • 30-day data retention
  • Standard support, 12x5

• Enterprise

  Custom

  annual commit, credited to usage

  • Unlimited clusters, multi-region and multi-cloud
  • SSO and SAML, custom RBAC
  • Self-hosted option
  • Premier support, 24x7

Plans as listed on the official pricing page on 24 Oct 2026. Prices can change and may vary by country or exclude tax. Check the official pricing page.

Union.ai alternatives

The main Union.ai alternatives listed on Debuly are Metaflow, Kubeflow, Anyscale, NVIDIA Dynamo-Triton, Together AI and Vast.ai. The table compares pricing and platforms; order is A–Z.

1. 

   Anyscale

   Managed platform for running Ray workloads, with pay-as-you-go compute and options to deploy in your own cloud.

   Developer platformsAutomation and agents
   PaidVisit Anyscale (official site)
2. 

   Kubeflow

   Open source, Kubernetes-native AI platform made of modular projects for training, tuning, pipelines, notebooks and model management.

   Developer platformsAutomation and agents
   Open sourceVisit Kubeflow (official site)
3. 

   Metaflow

   Open-source Python library for building ML, AI and data science workflows locally and running them at scale in your cloud.

   Developer platformsAutomation and agentsCoding
   Open sourceVisit Metaflow (official site)
4. 

   NVIDIA Dynamo-Triton

   Open-source inference server, formerly Triton Inference Server, for serving models from TensorRT, PyTorch, ONNX and more in production.

   Developer platforms
   Open sourceVisit NVIDIA Dynamo-Triton (official site)
5. 

   Together AI

   AI cloud for running open-source models through serverless and batch inference, dedicated endpoints, GPU clusters and fine-tuning.

   Developer platformsCoding
   PaidVisit Together AI (official site)
6. 

   Vast.ai

   GPU cloud marketplace for renting on-demand, interruptible or reserved GPUs with per-second billing.

   Developer platformsAutomation and agents
   PaidVisit Vast.ai (official site)

Union.ai FAQ

Sources & verification

Facts on this page come from the official sources below, each with the date we read it. Written by Debuly; the vendor did not review this page.

Image credits

• Logo: from cdn.prod.website-files.com · fetched 24 Oct 2026 · Vendor trademark, used only to identify the product; removed on request
• Cover: none collected yet, so a generated tile is shown.

Spot an error? Report an issue with this listing. We also remove a logo or image on the owner's request.