FICHA · MANJARO

onnxruntime-cuda

Cross-platform, high performance scoring engine for ML models (with CUDA)

  • Runtime
  • RUNTIME
  • MACHINE-LEARNING
  • ONNX
  • Dependency only
official+codex · reviewed · May 28, 2026 description in en

Description

Provides ONNX Runtime with CUDA support for machine learning model inference on compatible NVIDIA GPUs. GPU acceleration can improve throughput and latency for supported models.

Use it when the system has matching NVIDIA drivers, CUDA support, and enough GPU memory. Unsupported operators or driver mismatches can fall back, fail, or perform poorly.

Permissions

Permissions not analysed for this source yet.