UnionAI
Union.ai is an AI operation tool designed specifically to bring harmony to engineering, operations, and data science teams. It provides a platform for efficient...
Last verified:
What is UnionAI?
Union.ai is a production-grade platform for building, serving, and scaling complex AI systems, data workflows, and ML pipelines using pure Python. It powers dynamic workflows with real-time logic, conditions, retries, and fanout capabilities, enabling teams to ship fast from experiment to production without infrastructure burdens. Built on the open-source Flyte platform, Union.ai provides Kubernetes-native orchestration optimized for AI/ML, agentic workloads, and large-scale data processing across industries like biotech, autonomous systems, geospatial, and fintech.
Key features include infra-aware orchestration that dynamically provisions resources, output caching for efficiency, live remote debugging, task-level overrides, and integrated cost tracking. It supports ultra-low latency real-time inference, reusable containers, and automatic recovery from interruptions with 30-day data retention on the Team plan. Union.ai optimizes compute usage by charging only for allocated resources down to the second, supporting GPUs like H100 and B200, CPUs, and memory without idle capacity costs.
It's designed for AI/ML engineers, data scientists, and development teams at startups, enterprises, and organizations needing scalable, reliable AI delivery. Companies like Woven by Toyota, Rezo, LGND, and Spotify use it to cut costs by over 90%, scale to 160 GPUs, and accelerate R&D. Union.ai suits teams migrating from tools like Airflow, seeking multi-cloud support, and requiring enterprise-grade security with SSO, RBAC, and VPC options.
The platform emphasizes Python-native development, allowing local testing with the same SDK as production, and provides white-glove support for mission-critical applications. It enables massive fanout up to 50,000+ actions per run and custom concurrency, making it ideal for geospatial AI, drug discovery, autonomous driving, and agentic research.
UnionAI pricing
Pricing model: Paid
$950/mo Team plan includes $950 usage credit (min spend), 1,000 concurrent actions, 30-day retention, 1 cluster; Enterprise custom with volume discounts, custom concurrency/retention/clusters, 3+ clusters, white-glove support. Usage: GPU $0.1516-$2.8483/hr (T4g to B200), vCPU $0.0417/hr, Memory $0.0051/GB-hr; $0.0075/action on Team. No idle compute charges; discounts for startups/NGOs; BYOC supported.
UnionAI pros
- Pure Python orchestration for AI/ML workflows
- Scales to 160 GPUs seamlessly for geospatial AI
- Saves >90% on compute costs via caching
- Dynamic runtime decisions with conditions and retries
- Live remote debugging and task overrides
- Infra-aware provisioning reduces idle capacity
- Supports H100, B200 GPUs with per-second billing
- Migrates easily from Airflow with acceleration
- Fanout to 50,000+ actions per workflow run
- Real-time inference at ultra-low latency
- Kubernetes-native for multi-cloud environments
- Integrated cost tracking and observability
- Automatic recovery from interruptions
- Reusable containers across workflows
- Proven by Toyota, Spotify, Warner Bros
- Local testing with production SDK
- Volume discounts for high usage
- Bring-your-own-cloud deployment
UnionAI cons
- Minimum $950/mo spend on Team plan
- Usage-based billing can add up for heavy GPU use
- Requires Kubernetes cluster management
- Enterprise plan needed for custom retention
- Limited to 1 cluster on Team plan
- No free tier beyond potential trials
- Self-hosted control plane only on Enterprise
- Action concurrency capped at 1,000 on Team
- 30-day data retention limit on base plan
- Custom security like SAML on Enterprise only
Frequently asked questions about UnionAI
Do you offer discounts for startups?
Yes we do! Talk to our Startup Team to learn more.
Do you offer discounts for NGOs, universities, or non-profits?
Yes we do! Talk to our Public Sector Team to learn more.
How do monthly credits work for the Team plan?
Your monthly plan fee is issued back to you as usage credits. In practice, this means your monthly plan cost becomes your minimum monthly spend, and you can use that same amount in usage at no additional charge. Any usage beyond that amount is billed separately.
What is an action?
An action is an individual execution of a task. It represents a specific invocation of a task with particular inputs. If a task runs multiple times (such as inside a loop) you’ll see multiple actions, one for each invocation.
Can my team have a forward-deployed engineer (FDE) from Union.ai to help us build?
Yes, this is available as part of a Union Enterprise plan. Book a consultation with an engineer to learn more.
Do you offer a self-hosted control plane as a deployment option?
Yes, this is one of the deployment options available as part of a Union Enterprise plan. Self-hosted deployment supports on-prem, hybrid, and airgapped configurations. Book a consultation with an engineer to learn more.
How do you calculate GPU, CPU, Memory hours of usage?
We report the allocated resources (CPU, Memory, and GPU accelerator) from each container running the actions within your workflows and apply usage-based pricing down to the second. We do not include the resources consumed by any other services. Therefore, if you run Union on a shared K8s cluster, you are only paying for usage on the resources consumed by your Union tasks and workflows.
Is Union.ai a SaaS service?
No. You deploy the Union operator into a Kubernetes cluster you manage, which securely communicates with the Union control plane to poll for work. Your workflow executions, code, images, data, logs, and secrets all remain in your VPC/cloud, and are inaccessible to Union.
What’s the difference between action concurrency and actions/run (i.e., task fanout)?
Fanout is the total number of actions a run creates, while concurrency is how many of those actions are running at the same time. For example, a run might fan out to 50,000 actions but only execute around 100 of them concurrently.
Can I run Union.ai in my own cloud environment?
Yes, Union.ai supports bring-your-own-cloud (BYOC) deployments. You can run it in your own AWS, GCP, Azure, or neo-cloud environment while maintaining full control over your data, security, and infrastructure.