NVIDIA's DGX Spark AI systems are now receiving a brand new update that offers higher performance and stability.
NVIDIA OTA Update For DGX Spark System Brings Faster Performance, Better Stability, & Improved AI Support
NVIDIA's DGX Spark is a Mini AI Supercomputer that is now available on retail shelves. The cheapest systems start at $3144 US, and are available in NVIDIA's own reference design & partner offerings. We recently published an article on how the system delivers solid gaming performance, despite being an AI-first platform. With that said, NVIDIA is already rolling out the first OTA Update for DGX Spark systems, which offers a lot of improvements.
The latest OTA Update, which can be treated similarly to the GeForce Game Ready drivers for GeForce gaming GPUs, is aimed at the Spark system with the NVIDIA GB10 Superchip. This update offers improved performance, better stability, and enhances various workflows across the OS, GPU stack, Jupyter Lab, and connectivity.
The following are some of the enhancements that users can expect from the latest update:
What’s New:
- New DGX OS Kernel (Ubuntu 6.14 HWE stack): Delivers ongoing performance gains, tighter stability, broader hardware compatibility, and the latest security updates.
- Updated JupyterLab Environment: Now running CUDA 13.0.2 and the latest PyTorch stack for immediate access to updated frameworks without extra downloads.
- Accurate Memory Reporting: DGX Dashboard now aligns with CUDA unified memory guidance.
- Improved Peripheral Interoperability: Better compatibility with USB-C devices, monitors, Bluetooth peripherals, and Wi-Fi access points.
- Stable Diffusion XL Fix: Example workflow now runs end-to-end inside JupyterLab.
- Recovery Image Reliability: Installs correctly on macOS and with multiple external USB drives connected.
- Keyboard Accessibility Improvements: Smoother, more complete Out-of-Box Experience (OOBE) navigation.
This OTA update can be downloaded across all Spark platforms. In addition to this, NVIDIA is also actively working with 3rd party ecosystem partners to bring additional software support to the AI platform, such as Llama.cpp. This will enable improved memory management on unified-memory systems, ensuring accurate reporting of resources available on the platform.
Update Best Practices
- Use the DGX Dashboard: Always prefer the DGX Dashboard for system updates to ensure compatibility and optimal performance.
- Regular updates: Check for updates regularly, especially security patches
- Backup before major updates: Always back up critical data before major system changes
- Stable power: Ensure your system has a stable power supply during updates
- Maintenance windows: Schedule updates during planned maintenance windows when possible
To update your DGX Spark system, you can either navigate to the DGX Dashboard or input the respective command-line instructions to pull updates and upgrades for the system. NVIDIA has a useful guide here so you can follow that to manually update your DGX Spark system.
Follow Wccftech on Google to get more of our news coverage in your feeds.
