Nodes

TIR Nodes are fully collaborative environments that make AI development possible. They combine the power of containers, Jupyter Labs, and AI/ML frameworks to create a readily usable workspace for you and your entire team.

Getting Started

TIR Nodes

Guide To TIR Nodes

GPU and Performance

Using H100

Optimize workloads using H100 GPUs

Configuration and management

Committed Nodes

Use and manage committed nodes.

TIR-Provided Images

Pre-built images offered by TIR.

Customization and Integrations

Node with Your Own Container

Deploy nodes with a container of your choice.

Access Google Drive in Nodes

Integrate Google Drive with your TIR Nodes.

TIR Add Ons

Gradio

Create user interfaces

Tensorboard

Visualize machine learning training logs

TIR Guides

LLaMA Factory

Fine-tuning with LLaMA Factory on TIR

Volume Mount and Data persistence

Integration of Datasets (both EOS & Disk)

BuildKit

Build Docker Images on TIR Nodes Using a Dockerfile

Some of the most common use cases are:

• Run a script or notebook to fine-tune a Large Language Model (LLM) on a single GPU using PyTorch or Hugging Face train.
• Run a script or notebook to tokenize and fine-tune LLMs or Diffusion models with multiple GPUs (single machine) using DeepSpeed and Accelerate.
• Open and run a Jupyter notebook (.ipynb) from platforms like GitHub, Kaggle, or Colab.
• Download and review datasets stored on TIR or other platforms like Hugging Face.
• Download and test models like Stable Diffusion or any LLM.

Note

A TIR node is a fully functional coding environment. If you prefer to work with the command line (shell) over Jupyter Labs, you can configure SSH on a notebook (node). 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.