Linux support for NPUs is still evolving but making progress. Intel's NPUs can be used with their OpenVINO toolkit (works on Ubuntu/Fedora), and AMD has their ROCm platform for their NPUs. You can run stuff like Stable Diffusion locally using these - check out Automatic1111 which can utilize NPU acceleration. If you're looking for the best power efficiency for AI workloads, you might want to compare some power stations on gearscouts.com to keep your system running during longer inference sessions.
this post was submitted on 28 Jul 2025
7 points (100.0% liked)
Linux
12291 readers
60 users here now
Welcome to c/linux!
Welcome to our thriving Linux community! Whether you're a seasoned Linux enthusiast or just starting your journey, we're excited to have you here. Explore, learn, and collaborate with like-minded individuals who share a passion for open-source software and the endless possibilities it offers. Together, let's dive into the world of Linux and embrace the power of freedom, customization, and innovation. Enjoy your stay and feel free to join the vibrant discussions that await you!
Rules:
-
Stay on topic: Posts and discussions should be related to Linux, open source software, and related technologies.
-
Be respectful: Treat fellow community members with respect and courtesy.
-
Quality over quantity: Share informative and thought-provoking content.
-
No spam or self-promotion: Avoid excessive self-promotion or spamming.
-
No NSFW adult content
-
Follow general lemmy guidelines.
founded 2 years ago
MODERATORS
I believe some projects using OpenVino and ONNX can make use if the NPU. Maybe the generative AI plugins for GIMP and Audacity. But it's not a lot.
Most distros with have Arm and x86/64 versions so it is quite likely it can.
Not sure what the state of the Risk-v instruction set is.
But without knowing the device itself there is limited information we can give you.
This isn't an Arm vs x86 thing though, newer CPUs pretty much across the board from Intel/AMD/Qualcomm have an additional NPU that they use for AI acceleration.
As for support, you might be able to use them? It probably depends on the exact software but they might be able to be used for local LLMs. It's not exactly clear what practical uses they have even on Windows so that doubly applies to Linux just because it's more niche.