Pricing details
To suit all user needs, IBM Watson® Machine Learning has a variety of pricing plans, as well as cloud, server and local deployment options. Our plans give you the flexibility to deploy and run anywhere in a hybrid environment, helping you accelerate time to value for your data science projects. IBM Watson Machine Learning is also a key component for training and deploying AutoAI with IBM Watson Studio.
Which option is right for you?
Watson Machine Learning Cloud
The fastest way to move models from experimentations on your desktop or in your lab to deployments for production workloads. Runs in a managed service in a fully managed IBM Cloud™ environment.
Features:
- Pay-as-you-go consumption with as-a-service delivery, allowing you to be up and running in seconds
- Fully managed service, removing the need for infrastructure cost overhead
- Flexible interfaces available for managing models and configuring deployments
- End-to-end lifecycle support for machine learning, from training to deployment and evaluation
- First-class support for the most popular open source frameworks, as well as IBM SPSS®, IBM Decision Optimization for Watson Studio, and AutoAI with IBM Watson Studio
| Plan | Features | Pricing |
|---|---|---|
| Watson Machine Learning Cloud Lite |
Includes a maximum of five deployed models, 5,000 predictions per month, and 50 capacity unit hours per month for training or batch deployments. During this period, models can be trained, evaluated, and deployed to return model predictions. |
Free |
| Watson Machine Learning Cloud Standard |
Billed at a flat rate per thousand predictions and per capacity unit hour. Predictions are consumed by making REST calls to a deployed online model; capacity unit hours are consumed by executing training and batch scoring jobs. |
USD 0.50 /1,000 predictions USD 0.50 /capacity unit hour |
| Watson Machine Learning Cloud Professional |
Billed at a fixed amount per month, including 2,000,000 predictions and 1,000 capacity unit hours. Overage is billed at a flat rate per thousand predictions and per capacity unit hour. Predictions are consumed by making REST calls to a deployed online model; capacity unit hours are consumed by executing training and batch scoring jobs. |
USD 1,000.00 /month Overages: USD 0.40 /1,000 predictions USD 0.40 /capacity unit hour |
Watson Machine Learning Server
Lightweight machine learning server with simple installation on a single server and integration with Watson Studio Desktop for end-to-end data science. It is also designed to help you deploy your open source models from Jupyter notebooks or other tools with manageability and visibility.
Features:
- Single-server deployment option for simple installation and administration
- Integration with Watson Studio Desktop for model training and deployment
- Flexible interfaces available for managing models and configuring deployments
- End-to-end lifecycle support for machine leaning, from training to deployment and evaluation
- First-class support for the most popular open source frameworks, as well as SPSS
| Plan | Features | Pricing |
|---|---|---|
|
Watson Machine Learning Server – for small teams needing to scale compute and deploy models |
Term and perpetual licenses available. Deployed to a single node server in the private or public cloud of your choice, including AWS and Azure. |
Watson Machine Learning Local
All the benefits of Watson Machine Learning in the IBM Cloud, now deployed behind your firewall or in the public cloud of your choice.
Features:
- Can be deployed with Watson Studio Local
- Flexible interfaces available for managing models and configuring deployments
- End-to-end lifecycle support for machine learning, from training to deployment and evaluation
- Configurations available for HA and non-HA
- First-class support for the most popular open source frameworks, as well as SPSS and IBM Decision Optimization for Watson Studio
| Plan | Features | Pricing |
|---|---|---|
|
Watson Machine Learning Local – for enterprise data science teams |
Term and perpetual licenses available. Deployed in the private or public cloud of your choice, including AWS and Azure. Add-ons available: Watson Machine Learning Accelerator |