Getting Started

How to get started with Transformerengine

1

Verify System Requirements

Ensure your system runs Linux with CUDA 12.1 or higher (12.8+ for Blackwell GPUs), cuDNN 9.3+, Python 3.12 recommended, and an NVIDIA GPU with Compute Capability 8.9 or above for FP8 support.

2

Install Transformer Engine

Install via pip using the command: pip3 install --no-build-isolation git+https://github.com/NVIDIA/TransformerEngine.git@stable. The installer sets the NVTE_FRAMEWORK environment variable automatically if needed.

3

Import and Use in PyTorch

Import the library in your PyTorch code with: import transformer_engine.pytorch as te, then use modules such as te.Linear to build Transformer layers.

4

Run on CUDA Device

Execute your model on a CUDA device, for example by creating input tensors on the GPU: inp = torch.randn(..., device='cuda').

5

Refer to Documentation

Consult the Quickstart Notebook and official documentation for detailed examples and advanced usage.