Scalable, compliant & MLops-driven model deployments

User focused

Putting the machine learning engineer in the driver seat of a strongly MLOps focused model deployment flow.


Detailed resource, usage and performance monitoring out-of-the-box.


Everything you need to adhere to the newest regulations around AI deployment.

User focused

  • Lightweight integration

    Preparing models is as easy as adding a few lines of instructions in your code.

  • Freedom of choice

    Due to our clean process of deployment you will retain maximum freedom in choosing the modelling tools you deem relevant.

  • MLOps focused

    Our CLI & UI combination is highly intuitive to use and allows for clean integration in your MLOps deployment pipelines.

Robust & scalable

  • Autoscaling

    Cost-efficient autoscaling of model instances from 0 to whatever fits your need.

  • High-availability predictions

    Be able to depend on your solutions whenever you need them.

Regulatory compliant

  • Monitoring

    Actively monitor model performance & usage, get actively notified when drift occurs.

  • Full-cycle prediction tracing

    You're able to understand how a prediction came to be even years after it was made.

  • Governed model store

    Transparent, persistent and comparable model versions.

Our customers trust us.

We are trusted to host millions of predictions per day.

12+ million
Predictions per day

Bring your framework
of choice

We make sure you can deploy it ❤️

Plans that scale with you

You monthly pricing depends on actual resource consumption. Automated cost-optimization is done for you based on scale bounds you defined.
Per Instance, Per Hour
  • 3GB RAM
  • 1 CPU core
  • Auto scale to 0
  • Scale to 3  instances / model
  • Unlimited storage & versions
  • For non-commercial usage
Try now
Perfect starter
Per Instance, Per Hour
  • Includes all free features
  • 4GB RAM
  • 1 CPU core
  • Free when not used
  • Auditable predictions
Get in touch
Highly customizable machine types
  • Includes all starter features
  • Up to 265+ GB RAM
  • Up to 64+ CPU cores
  • Up to 10+ GPUs
Get in touch
Can I upgrade later on?

Changing to different machine types is a easy as adjusting your deployment configuration.

Are all features included?

With the exception of single-sign on all features are available in the free version. Single-sign on is available from the moment you upgrade to a paid plan.

Is "free" completely free?

It is! As long as the free machine type suites your needs you're good to go. Are you hosting a data-for-good initiative and looking for a sponsor, contact us!

Can I request high-performance SLAs?

Sure, just get in touch with your exact questions and we'll get back to you ASAP!

Still have unanswered questions? Get in touch

Get started today, no credit card required.

Get in touch