Skip to main content

Launch GPU/CPU powered instances

Launching an instance is super simple and flexible 😊. Choose the framework, GPU type, GPU quantity and storage. Click the launch button. Your instance should be up before you finish scrolling your favourite deeplearning paper, in less than 30 seconds ⏰. Launch

Detailed walk through​

You can launch an instance in 4 easy steps. Choose

  • Framework

    • Pytorch
    • FastAI
    • TensorFlow
    • BYOC (Choose any public container of choice - Advanced Feature)
  • GPU Type in ascending order of performance.

    • RTX 5000
    • RTX 6000
    • A5000
    • A6000
    • A100
    • CPU
  • Number of GPUs

    • Choose a value between 1 to 8
  • Choose disk size

    • Choose a value between 20GB to 500GB.
  • Reserved

    • Yes
    • No
  • Duration

    • Hourly
    • Weekly
    • Monthly

Reserved vs Not reserved​

We provide 2 kinds of instances.

  • Reserved
  • Not reserved


When you choose reserved instances, the resources are guaranteed. The resources that are allocated, will not be realease until you hit pause/destroy or run out of balance.

Not Reserved or Spot instances​

Instances which are not reserved are called as spot or preemptible instances. You can use spare GPU capacity at discounted prices and save upto 60% cost. But these instances wil be auto-paused when demand increases. These kind of instances are suited for

  • Learning.
  • When you have implemented checkpoints, and you want to save on costs.

Weekly/Monthly duration​

Get upto 10% discounts for weekly usage and upto 30% discounts on monthly usage. Instances that are launched for a week/month cannot be paused or destroyed during that duration. Post the duration, instances will continue at the discounted prices till you choose to pause/destroy.

Optional features​

Click on the down arrow ⬇️ for additional options.

Startup script πŸ“œβ€‹ Startup Script

You can choose a starup script and it gets executed on start of the instance, you can use it do things like

  • Install your favourite libraries.
  • Clone github repos.

Name the instance​ Name Instance

It is common to run multiple experiments while training models. Name the instance to remember what project you are using it for.


Billing starts once an instance is created, so ensure to either pause/destoy the instance when not in use to avoid un-neccessary costs.

Take care

The instances will be paused automatically, if the account balance becomes negative. You can enable subscription or maintain sufficient credits in the wallet.