DGX Spark Release Notes#

This section provides release notes for the DGX Spark, including information about new features, known issues, and software version updates.

Current Software Versions#

The following table shows the current version information for the DGX Spark software stack.

New DGX OS kernel - The operating system now incorporates the Ubuntu 6.14 Hardware Enablement (HWE) kernel stack. The new kernel brings ongoing performance gains, enhanced stability, broader hardware compatibility, and the latest security updates for a more secure operating environment.
JupyterLab updated to latest CUDA and PyTorch - Updated JupyterLab to CUDA 13.0.2 and the latest PyTorch version, enabling users to work with updated frameworks immediately without extra downloads.

Improved memory reporting in DGX Dashboard - Addressed the issue with memory reporting differences with unified memory architecture; readout now consistent with CUDA guidance for unified memory systems. See https://nvidia.custhelp.com/app/answers/detail/a_id/5728.
Image Generation in JupyterLab - Resolved the inability to generate images in Stable Diffusion XL Playbook example. Users can now run end-to-end example workflow inside JupyterLab.
Improved Peripheral Interoperability - Better compatibility with USB-C devices, monitors, Bluetooth peripherals, and Wi-Fi access points.
Improved recovery image reliability - The recovery image now installs correctly on macOS and also when multiple external USB-C drives are connected.
Keyboard accessibility improvements - Smoother and more complete out-of-the-box experience navigation, including the ability to complete entire system setup process using only keyboard.

For an updated list of known issues, see Known Issues.