CUDA DL Release 25.05

The NVIDIA container image for CUDA Deep Learning, release 25.05, is available on NGC.

.

Contents of the CUDA DL container

This container image contains the complete source of the version of PyTorch in /opt/pytorch. It is prebuilt and installed in the default Python environment (/usr/local/lib/python3.10/dist-packages/torch) in the container image.

The container also includes the following:

• Ubuntu 24.04 including Python 3.12
• NVIDIA CUDA 12.9.0
• NVIDIA cuBLAS 12.9.0
• NVIDIA cuDNN 9.10.1.3
• NVIDIA NCCL 2.26.5
• rdma-core 50.0
• NVIDIA HPC-X 2.23
• GDRCopy 2.4.1
• TensorBoard 2.16.2
• Nsight Compute 2025.2.0.11
• Nsight Systems 2025.3.1.90
• NVIDIA TensorRT™ 10.10.0.31

Driver Requirements

Release 25.05 is based on CUDA 12.9.0. For comprehensive and up-to-date driver compatibility information, please refer to the following documentation:

• NVIDIA CUDA Compatibility Guide - Compati.bility information between CUDA versions and driver releases.
• CUDA Toolkit Release Notes - Driver version requirements and compatibility matrices.
• NVIDIA Drivers Download - Latest NVIDIA drivers.

Key Features and Enhancements

This CUDA DL release includes the following key features and enhancements.

• CUDA DL container image is based on Ubuntu 24.04.

Announcements

• Starting with the 25.03 release, NVIDIA provides two CUDA DL images: Developer and Runtime

Known Issues

• None.