NVIDIA Brev#

Overview#
NVIDIA Brev is a user-friendly platform for launching cloud instances.
Brev (https://developer.nvidia.com/brev) is a paid service to launch and manage GPU instances from different cloud providers
It integrates with Workbench to simplify creating remote locations
Connecting Brev to Workbench is straightforward.
Configure the Brev integration in the AI Workbench Settings page
Create a Brev instance using the Brev console
Create a remote Location in AI Workbench and select the Brev instance you created
Using Brev with AI Workbench#
Prerequisites#
You must have AI Workbench installed locally
You must have a valid email address to create an NVIDIA account
Step 1: Configure integration#
This is a one time setup that requires NVIDIA account creation if you don’t yet have one.
If you already have an NVIDIA account, then use that email when connecting the Brev integration.

Go to the AI Workbench Settings page
Click “Connect” on the Brev card
Enter your email when prompted, and follow the account creation steps if needed
Wait until the integration and authentication are confirmed in your browser
Note
There are a few steps if you are creating an NVIDIA account for the first time:
Receive a confirmation email with a code needed to initiate account creation
Create your account and password
Create your NVIDIA Cloud Account and give it a name
Log in to your NVIDIA account to finalize the Brev integration
Step 2: Create an instance in Brev#
You must create an instance on Brev before you can add it as a Workbench ___location.

Log in to Brev at brev.nvidia.com, go to the GPU tab, and click Create New Instance
Select the GPU type, and then select the configuration and provider
Configure the disk size, give the instance a unique name, and click Deploy to launch the instance
Warning
The AI Workbench integration with Brev does not support Lambda Labs because their instances lack start/stop capabilities.
You can use any other cloud provider available in Brev, including AWS, Azure, Crusoe and GCP.
Step 3: Add the Location#
Once the integration is connected, you can add running Brev instances directly in the AI Workbench remote ___location modal.

Go to the Locations page in the Desktop App
Select Add Remote Location at the top right
Select Create Brev Location and click Next
Select the Brev org from the drop down at top right
Select the instance you created and click Add Location
Step 4: Work in the Location#
The Brev ___location is like any other remote Workbench ___location with the added benefit that Workbench will start/stop the instance when you open/close the ___location.
Opening the Brev ___location in the Desktop App will start the instance, and depending on the cloud provider this can take a few minutes
Closing all AI Workbench windows connected to the Brev ___location will stop the instance
Warning
You should always double-check the Brev console (brev.nvidia.com) to confirm instances are shutdown.
FAQs#
Is Brev a cloud provider?#
No. Brev partners with major cloud providers to offer a single platform to deploy and manage GPU instances.
Is Brev free?#
No. Brev has a flexible, pay-as-you-go model where you only pay for the time you use an instance. You need a credit card to provision an instance.
Can I configure the Brev integration with the CLI?#
Yes. You can configure the integration and you can create a Brev ___location using the CLI.
Use the command nvwb connect integration
and then select Brev from the drop down.
You can add a running Brev instance with the command nvwb create context --brev
. Select the
instance name from the drop down.
Why doesn’t my Brev instance show up in Workbench?#
There are a few potential reasons:
You haven’t configured the Brev integration with that particular account
You already created a ___location with that instance
You are looking in the wrong Brev org in the remote ___location modal
I see a Brev instance in the remote ___location modal, but I can’t add it as a ___location. Why is that?#
There are a few potential reasons:
The instance is not running
The instance is a Lambda Labs instance and AI Workbench does not support using them as a remote ___location