Instances (Nodes)

Instances provide collaborative, ready-to-use environments for AI and ML development. Each instance brings together containers, JupyterLab, and popular AI frameworks—giving you and your team a powerful workspace that works out of the box.

Getting Started

Instances(Nodes)

Guide to Instances

Managing Instance

Manage Instances

Customization and Integrations

Instance with Your Own Container

Deploy Instances with a container of your choice

Access Google Drive in Instances

Integrate Google Drive with your Instances(Nodes)

Add Ons

Gradio

Create user interfaces

Tensorboard

Visualize machine learning training logs

Guides

LLaMA Factory

Fine-tuning with LLaMA Factory

Storage Mount and Data persistence

Integration of Datasets (both EOS & Disk)

BuildKit

Build Docker Images on Instances(Nodes) using a Dockerfile

What you can do with Instances

• Fine-tune a model on a GPU.
• Train large models across multiple GPUs.
• Open and run Jupyter notebooks from GitHub, Kaggle, or Colab.
• Work with datasets stored on our platform or from external sources like Hugging Face.
• Try out models such as Stable Diffusion or popular LLMs.

Note

Instance (Node) is a fully functional coding environment. If you prefer to work with the command line (shell) over Jupyter Labs, you can configure SSH. This way, you can upload your data using SFTP or sync your code with Git tools and run the scripts as you would on your local system.