Sold by: Union.ai
Deployed on AWS
Union.ai delivers 9.8x ROI as the Durable AI Runtime Engine powered by Flyte.org. Automatic fault recovery, native spot instance scaling, and self-healing execution for agentic AI, distributed training, and real-time inference. Write pure Python, declare your compute requirements, and run fully managed on Amazon EKS in your AWS VPC. Trusted by Woven by Toyota, Artera and Rezo.
4.5
Overview
Video
Union.ai Product Tour
Video
Product video
Union.ai is the Durable AI Runtime Engine built by the creators of Flyte the open-source orchestration platform trusted by over 3,000 companies. Powered by Flyte 2, it provides a pure Python execution model that embeds durability directly into your AI workflows. Agentic systems, training jobs, and inference pipelines automatically recover from infrastructure failures without losing state, no custom recovery logic, no lost work, no restarts from scratch.
Union.ai runs entirely in your AWS VPC via Amazon EKS. Your data, code, and model artifacts never leave your infrastructure, we have Zero Trust, are SOC 2 Type II certified and HIPAA compliant. Enterprise SSO, RBAC, and full audit trails included.
New to Union.ai? Start with FlyteOS the open-source, single-cluster deployment of Flyte 2, now generally available. Zero software charge. Deploy in minutes via Flyte DevBox. Visit flyte.org to setup your intial validation process.
Highlights
- Runs natively on AWS in your VPC. Union.ai provisions and manages a Kubernetes cluster entirely inside your AWS VPC via Amazon EKS. Your data plane - including all data, code, model artifacts, and compute - remains exclusively within your infrastructure at all times. Native integrations with Amazon Bedrock, S3, SageMaker, Athena, CloudWatch, Trainium, and Inferentia. Zero-data-exfiltration architecture by design.
- Build self-healing AI/ML workflows in minutes with a pure Python execution model that embeds durability directly into your code. Agents, training jobs, and inference pipelines automatically recover from infrastructure failures without losing state, delivering analyst-validated 9.8x ROI.
- Power agentic AI with durable execution. Native integrations with OpenAI, Claude, Mistral, LangChain, LangGraph, CrewAI, Google ADK, and Pydantic AI. Handle loops, branching, and real-time decision-making using standard Python. When tool calls or compute nodes crash, workflows self-heal and resume exactly where they left off.
Details
Sold by
Delivery method
Deployed on AWS
New
Introducing multi-product solutions
You can now purchase comprehensive solutions tailored to use cases and industries.
Features and programs
Financing for AWS Marketplace purchases
AWS Marketplace now accepts line of credit payments through the PNC Vendor Finance program. This program is available to select AWS customers in the US, excluding NV, NC, ND, TN, & VT.
Pricing
Pricing is based on the duration and terms of your contract with the vendor, and additional usage. You pay upfront or in installments according to your contract terms with the vendor. This entitles you to a specified quantity of use for the contract duration. Usage-based pricing is in effect for overages or additional usage not covered in the contract. These charges are applied on top of the contract price. If you choose not to renew or replace your contract before the contract end date, access to your entitlements will expire.
Additional AWS infrastructure costs may apply. Use the AWS Pricing Calculator to estimate your infrastructure costs.
Dimension | Description | Cost/12 months |
|---|---|---|
Annual Commit | Fee for Union.ai, credit for monthly usage based on tier | $100,000.00 |
Dimension | Description | Cost/unit |
|---|---|---|
Usage in Union.ai beyond the monthly credit | Extra usage beyond monthly credit based on actions and action-allocated resources | $0.01 |
Dimensions summary
You pay through two connected dimensions. The Annual Commit is a yearly fee that also gives you a monthly usage credit tied to your tier. That credit becomes your minimum monthly spend, and you can use up to that amount at no extra charge. The second dimension covers usage beyond the monthly credit. It bills based on the actions you run and the resources (CPU, memory, GPU) allocated to those actions, measured down to the second. Together, these dimensions set a predictable base cost plus variable charges for work that exceeds your included credit.
Top-of-mind questions for buyers
An action is one unit of work: an invocation of a task, trace, or condition that runs end to end to completion. Duration does not matter, whether it finishes in seconds or runs for a month. Retries do not add actions; a task counts once it succeeds. Status and health checks also do not count.
You are billed for the resources allocated to each container running your actions, measured down to the second. Only resources consumed by your own tasks and workflows count. If you run on a shared cluster, other services on that cluster are not included in your charges.
Your Annual Commit issues a monthly usage credit tied to your tier. That credit becomes your minimum monthly spend, and you use up to that amount at no extra charge. Only usage above the credit triggers charges under the second dimension, based on actions and allocated resources.
union.ai
Helpful?
Vendor refund policy
All fees are non-cancellable and non-refundable except as required by law.
Custom pricing options
Request a private offer to receive a custom quote.
Legal
Vendor terms and conditions
Upon subscribing to this product, you must acknowledge and agree to the terms and conditions outlined in the vendor's End User License Agreement (EULA) .
Content disclaimer
Vendors are responsible for their product descriptions and other product content. AWS does not warrant that vendors' product descriptions or other product content are accurate, complete, reliable, current, or error-free.
Delivery details
Software as a Service (SaaS)
SaaS delivers cloud-based software applications directly to customers over the internet. You can access these applications through a subscription model. You will pay recurring monthly usage fees through your AWS bill, while AWS handles deployment and infrastructure management, ensuring scalability, reliability, and seamless integration with other AWS services.
Resources
Vendor resources
Support
Vendor support
Enterprise customers receive a dedicated Customer Success Manager, direct Slack access to Union.ai engineering, SLA-backed support, and architecture review assistance. For partner or sales inquiries: info@union.ai
AWS infrastructure support
AWS Support is a one-on-one, fast-response support channel that is staffed 24x7x365 with experienced and technical support engineers. The service helps customers of all sizes and technical abilities to successfully utilize the products and features provided by Amazon Web Services.
Updated weekly
Sentiment is AI generated from actual customer reviews on AWS and G2
Functionality
Ease of use
Customer service
Cost effectiveness
Positive reviews
Mixed reviews
Negative reviews
AI generated from product descriptions
Kubernetes-based Infrastructure Management
Provisions and manages Kubernetes clusters entirely within AWS VPC via Amazon EKS with native integrations to Amazon Bedrock, S3, SageMaker, Athena, CloudWatch, Trainium, and Inferentia.
Automatic Fault Recovery and State Preservation
Automatically recovers from infrastructure failures without losing state through embedded durability in workflows, enabling agentic systems, training jobs, and inference pipelines to resume execution from the point of failure.
Pure Python Execution Model
Supports pure Python-based workflow development with declarative compute requirements, enabling direct embedding of durability logic without custom recovery implementations.
Agentic AI Framework Integration
Provides native integrations with OpenAI, Claude, Mistral, LangChain, LangGraph, CrewAI, Google ADK, and Pydantic AI for handling loops, branching, and real-time decision-making.
Enterprise Security and Compliance
Implements Zero Trust architecture, SOC 2 Type II certification, HIPAA compliance, Enterprise SSO, RBAC, and full audit trails with zero-data-exfiltration design ensuring data, code, and model artifacts remain within customer infrastructure.
Workflow Orchestration
Orchestrates and automates data workflows with real-time monitoring and execution capabilities across distributed environments.
Real-time Observability
Provides real-time visibility and monitoring across the entire data stack with workflow visualization and performance tracking.
Python-native Development
Enables workflow development directly in Python, allowing developers to build data pipelines using native Python code without requiring domain-specific languages.
ETL/ELT Pipeline Automation
Automates extract, transform, and load operations with real-time monitoring, error recovery, and data integration capabilities.
Security and Governance Framework
Enforces data privacy, security controls, and governance frameworks throughout data workflow execution and management.
Multi-Language Support
Supports Python, R, SQL, GraphQL, PySpark, and Markdown blocks with full interoperability between languages in the same project.
Flexible Compute Options
Provisioning of compute resources per task on demand including Lambda, Fargate, GPU, and Kubernetes with automatic release after task completion.
Native Data Connectors
Built-in connectors for Snowflake, PostgreSQL, MySQL, MariaDB, Hugging Face, and AWS Bedrock for direct integration without additional setup.
Role-Based Access Control and Security
RBAC, SSO, granular permissions, encrypted credential vault, and self-hosted deployment through CloudFormation with data and execution outputs stored in customer AWS account.
Real-Time Collaborative Development
Multiple users can simultaneously write and execute code in the same project with live change visibility, inline commenting, and GitHub integration for version control.
Standard contract
No
Customer reviews
Md Masudul Hassan
Modern AI workflows have become reproducible and visible for managing complex student projects
Reviewed on May 31, 2026
Review provided by PeerSpot
What is our primary use case?
My main use case for Union Cloud is to orchestrate machine learning workflows, automated data pipelines, and manage the AI development process across the cloud environment day-to-day.
A specific example of how I use Union Cloud in one of my recent projects is that it has significantly improved the way we manage AI and machine learning workflows, especially in student projects, such as when a student uses it for a research project.
What is most valuable?
Union Cloud helps our team move AI projects from experimentation to production with better reliability and scalability, and it enhances how the students or my team interact with it.
The best features that Union Cloud offers include strong AI and ML workflow orchestration, reliable workflow execution with automatic recovery, and caching, which stand out the most to me.
Union Cloud has positively impacted my organization by providing the ability to manage complex AI workflows while maintaining reliability and reproducibility. Reproducibility is one of the platform's biggest strengths. It also provides strong workflow visibility and debugging capabilities, which helps teams resolve issues quickly.
What needs improvement?
Union Cloud can improve by offering more low-code capabilities for non-technical users and expanded customization options for the workflow monitoring dashboard. I also wish for simplified configuration for smaller teams, as this is an area where the platform can improve.
For how long have I used the solution?
I have been using Union Cloud for more than three months.
What do I think about the stability of the solution?
Union Cloud is definitely stable in my experience.
What do I think about the scalability of the solution?
Regarding Union Cloud's AI capabilities, I believe its scalability is one of its strongest advantages. As workloads grow, the platform can efficiently scale training orchestration and inference processes across the cloud infrastructure, which makes it well-suited for organizations handling large-scale AI and data workloads.
From my understanding, Union Cloud's scalability is approximately nine.
How are customer service and support?
The customer support for Union Cloud, as I discussed with my team, is very positive. The team is knowledgeable, responsive, and focused on helping customers successfully deploy and optimize workflows.
What other advice do I have?
My advice to others looking into using Union Cloud is that it is well-suited for organizations handling large-scale AI or data flows, and also for teams who want to deploy any AI-related project.
Union Cloud is a modern AI orchestration platform that simplifies workflow management, improves stability, and accelerates the path from experimentation to production. I would rate this solution a nine out of ten.