Lambda GPU Cloud

Documentation for Lambda GPU Cloud, the #1 cloud provider for high-performance GPUs

Lambda GPU Cloud provides instant access to high-performance cloud GPUs at the best prices on the market.

Note

Beginning December 13, 2023, new Lambda GPU Cloud instances will launch with Ubuntu 22.04 instead of Ubuntu 20.04. Currently running instances won’t be affected by this change.

Significantly, Ubuntu 22.04 includes Python 3.10, while Ubuntu 20.04 includes Python 3.8.

See the release notes to learn more about the changes introduced in Ubuntu 22.04.

Looking to use our new API?

Read our documentation

With Lambda GPU Cloud, you have:

TensorFlow, JupyterLab, PyTorch®, and other popular ML software pre-installed
persistent storage to save your datasets and other files
root access to your instances via SSH

Are API requests rate-limited?

Can I access my persistent storage without an instance?

Can I launch an instance from the command line?

Can I list my running instances from a command line?

Can I list the offered instance types from a command line?

Can I pause my instance instead of terminating it?

Can I remotely mount persistent storage?

Can I set a limit (quota) on my file system usage?

Can I use the Cloud API to add an SSH key to my account?

Can my data be recovered once I've terminated my instance?

Can you provide an estimate of how much a job will cost?

Do you support Kubernetes (K8s)?

How are on-demand instances invoiced?

How are persistent storage file systems billed?

How do I change my billing address?

How do I change my password?

How do I get started using the dashboard?

How do I get started using the Demos feature?

How do I get started using the Firewall feature?

How do I get started using the Team feature?

How do I import an SSH key from a GitHub account?

How do I learn my instance's private IP address and other info?

How do I list my file systems using the Cloud API?

How do I open Jupyter Notebook on my instance?

How do I restart an instance using the Cloud API?

How do I retrieve the details of an instance from a command line?

How do I terminate an instance using the Cloud API?

How do I use persistent storage to save datasets and system state?

How long does it take for instances to launch?

Is it possible to open ports other than for SSH?

Is it possible to use more than one SSH key?

What can I do with the Cloud API?

What happens to my account if I don't pay an invoice?

What is the capacity of persistent storage file systems?

What network bandwidth does Lambda GPU Cloud provide?

What should I do about timeout waiting for RPC from GSP errors?

What SSH key formats are supported?

Why am I seeing an error about NMI received for unknown reason?

Why are some instance types grayed out when I try to launch an instance?

Why can't my program find the NVIDIA cuDNN library?

Why is my card being declined?

Feedback

Was this page helpful?

Glad to hear that! Contact us with any suggestions on how we can improve our docs.

Sorry to hear that. Please contact Support for help.

Last modified December 13, 2023: Make note more prominent (54be2ff)