Optimum

Overview

Conceptual guides

Nvidia

AMD

Ryzen AI

Brevitas

Intel

AWS Trainium/Inferentia

Google TPUs

Habana

Furiosa

ONNX Runtime

Exporters

BetterTransformer

Torch FX

LLM quantization

Utilities

You are viewing v1.21.2 version. A newer version v1.23.3 is available.

Join the Hugging Face community

and get access to the augmented documentation experience

Collaborate on models, datasets and Spaces

Faster examples with accelerated inference

Switch between documentation themes

to get started

🤗 Optimum-AMD

🤗 Optimum-AMD is the interface between the 🤗 Hugging Face libraries and AMD ROCm stack and AMD Ryzen AI.

In-depth guides and tools to use Hugging Face libraries efficiently on AMD GPUs.

Use pre-optimized models for AMD Ryzen AI NPU.