Red Hat's Christian Schaller shared an update on the continued enhancements to Fedora Workstation on Friday: Given current industry trends, ongoing Fedora Workstation development has been getting a lot of attention centered around AI, AI, AI.
Blog posts from Christian Schaller, Red Hat's Director of Desktop and Beyond, are always interesting as to what areas Fedora will be focusing on further in the future: long-standing goals remain, such as improving Linux desktop HDR support, but more recently there has been a lot of user/customer interest in moving forward with AI.
Highlights from Christian's Fedora Workstation development update include:
– Red Hat is committed to helping customers get more out of AI, and in Fedora, this is being achieved in part through initiatives such as IBM's Granite AI model, which is fully open, and InstructLab, which is beginning to leverage the Granite LLM model to teach new things.
– Fedora is also working to ensure other large language models like Meta's Llama run well. Fedora continues to work to make AMD ROCm accelerated PyTorch available in Fedora. In the long term, Fedora wants to make it easy to run PyTorch on hardware that can take advantage of CPU/GPU acceleration from all major vendors.
– Work continues to improve the experience of NVIDIA proprietary drivers on Fedora and make setup easier.
– Partly related to NVIDIA's efforts, Fedora by F41 Workstation wants to make it easier to self-sign kernel modules for use with Secure Boot.
– Toolbx container tools have improved integration of NVIDIA drivers.
– We continue to improve the Linux desktop's support for High Dynamic Range (HDR) displays.
– We also continue to work on improving how DRM leases handle VR headsets.
– Explicit sync support has been nicely integrated from the Wayland desktop/compositor to the NVIDIA and Mesa drivers and other parts of the stack.
For more information on Fedora Workstation development efforts, see Christian's blog.
