Templates on Inference.ai make it easy to launch powerful servers with everything you need already configured. Whether you’re training a large model, running experiments, or just getting started, templates save time and reduce setup errors.

Think of a template as a blueprint for your server: it includes the machine specs, installed software, storage volumes, and even environment variables, so you can get up and running with a single click.

In your account, there is one available template, which includes:

• Jupyter Notebook integration

• 1x GPU with 16GB VRAM

• 4x vCPU and 16GB RAM

This template has been pre-configured and provided by your Administrator for ease of use.

How to Use a Template

1. Navigate to the Server tab in the left sidebar

2. Select Create New Server

3. Select which template you would like to use

Why use Templates?

Templates are great for:

• Quickly spinning up repeatable environments

• Standardizing setups across team members

• Avoiding misconfiguration

• Saving time on recurring workflows (e.g., model training, grading scripts, class labs)

With templates, you don’t have to manually select your GPU, disk size, or volumes every time—you just launch and go.

Can I Create My Own Template?

At the moment, users cannot create their own templates, including admins. All available templates on Inference.ai are pre-configured by our team to support the most common use cases like training models, classroom environments, and quick-start AI workflows.

We're actively working on releasing self-serve template creation, which will allow organizations to define and share their own presets. Until then, you can browse and launch from the set of templates available to your organization.

Need a custom template? Reach out to [email protected] with your requirements, and we may be able to configure one for you.