Do not waste your money,
Pay as You Make Prediction

Modelify only charges when you run your model. It does not charge you any fees when you do not use the model.

Plans & Pricing

We only charge you based on the prediction duration of your model.

Free

no credit card needed

  • 1000 Request Free / month
  • 1 Application
  • Up to 5 instances
  • Up to 3 Model Versions
  • Logging
  • Monitoring
  • Up to 2 GB Memory
  • 1 vCPU
  • HTTPS API

Standard

pay as you go

  • Unlimited Request
  • Up to 20 Applications
  • Up to 100 instances
  • Unlimited Model Versions
  • Monitoring & Logging
  • Up to 32 GB Memory
  • Up to 4 vCPU
  • HTTPS API
  • Private API Support
  • Custom Domain
  • Workflow Tool

Enterprise

Contact Us

  • Unlimited Request
  • Unlimited Applications
  • Up to 1000 instances
  • Unlimited Model Versions
  • Monitoring & Logging
  • Up to 64 GB Memory
  • Up to 8 vCPU
  • HTTPS API
  • Private API Support
  • Custom Domain
  • Workflow Tool
  • Private Cloud Deployment
  • Multi-zone Replication
  • Professional Support

Billable prediction time

Modelify only bills you for the resources you utilise on prediction, which are rounded up to the nearest 100 millisecond. Modelify does not charge you for resources spent while the instance is idle and during shutdown-start times, as well as resource spent on requests outside of the model pipeline. In the figure below, you can see when you were billed.

how-modelify-billing

Modelify charges you for the resources you use when you use them. Your bill is calculated over 2 sources, CPU and Memory cost. The following table describes resource pricing.

CPU

The number of vCpu allocated to each instance of your application.

$0.00012

vCPU-second

Memory

The amount of memory (RAM) allocated to each instance of your application.

$0.000075

GiB-second

Helping you with machine learning operations and integrations

© 2021. All rights reserved. modelify.ai