Glossary

Key terms and concepts in GPUX.

A

ONNX - Open Neural Network Exchange, a standard format for ML models.

E

Execution Provider - Backend that runs inference (CUDA, CoreML, ROCm, etc.).

G

GPU - Graphics Processing Unit, hardware accelerator for ML.

GPUX - Docker-like GPU runtime for ML inference.

I

Inference - Running predictions on a trained ML model.

M

Model - Trained machine learning model in ONNX format.

P

Provider - See Execution Provider.

R

Runtime - The GPUX execution environment.

Tip

Confused by a term? Ask on Discord!