{"slug": "beyond-tokens-per-watt-using-ubuntu-26-04-lts-for-ai", "title": "Beyond tokens per watt – using Ubuntu 26.04 LTS for AI", "summary": "Canonical announced that Ubuntu 26.04 LTS will include single-command installations for NVIDIA CUDA, AMD ROCm, and NVIDIA DOCA-OFED, aiming to reduce deployment friction for AI infrastructure. The company is also partnering with silicon vendors to optimize the operating system for AI workloads, minimizing OS overhead to maximize hardware efficiency. The release targets enterprise needs for faster time-to-value and improved human productivity in managing AI stacks.", "body_md": "# Beyond tokens per watt – using Ubuntu 26.04 LTS for AI\n\n###\n[Freyja Cooper](/blog/author/freyja-cooper)\n\non 5 June 2026\n\n**Tags:**\n[AI/ML](/blog/tag/ai-ml)\n,\n[AI/ML Infrastructure](/blog/tag/ai-ml-infrastructure)\n\nTokens per watt (TpW) – the measure of useful AI work produced per watt of energy consumed – is the metric at top of mind for CEOs, heads of AI, and infrastructure teams alike. With the tremendous cost of GPU clusters, extracting as much value as possible from the expense is critical.\n\nBut in the pursuit of tokens, it’s important to remember that hardware efficiency isn’t the only factor influencing data center operating costs, or the output of useful, revenue-generating AI work. While TpW is crucial, we also need to consider time-to-value and the impact of human productivity, which are largely determined at the software level.\n\nWe’re shaping Ubuntu to be the software foundation for efficient AI, and in this article, I’ll share some examples of what we mean when we say that we are optimizing Ubuntu for AI. With Ubuntu 26.04 LTS, we’re not just helping organizations get more from their hardware, we’re also making life easier and more productive for teams that rely on and support the AI stack.\n\n## An OS that’s optimized for silicon\n\nHow do you squeeze more tokens from your hardware? The prevailing wisdom is to prioritize model optimization, GPU utilization, time to first token, and tokens per second. However, it’s also essential to have a software layer that enables you to make the most of your silicon.\n\nThe host operating system plays a central role in the AI infrastructure stack. That’s “central” not just in the sense that it’s important, but also in the sense that it sits in the center of the stack, acting as the bridge between the hardware and software. The OS manages the underlying compute, so it’s responsible for ensuring you can take full advantage of your GPUs and other AI accelerators.\n\nWith that in mind, Canonical [partners with silicon vendors](https://canonical.com/partners/silicon) (such as NVIDIA, AMD, Intel, Arm, and Qualcomm, as well as RISC-V platforms) to optimize Ubuntu across all major architectures. This optimization helps to ensure that the maximum watts are spent on AI workloads rather than OS overhead.\n\nWe also work with partners to [certify hardware](https://ubuntu.com/certified). By providing standardized, pre-integrated secure boot enablement and firmware delivery, Canonical enables organizations to avoid having to do custom OS engineering for every new piece of hardware they add to their stack. Enterprises can get to value faster, and save on engineering resources.\n\n## Single command toolkit integrations\n\nLet’s continue on that theme of accelerating time-to-value and enhancing human productivity. Even in the age of AI, Ubuntu remains a Linux for human beings, and a core pillar of our philosophy is minimizing the friction involved in deploying and operating AI infrastructure for our users.\n\nTo that end, we’re collaborating with NVIDIA and AMD to integrate and distribute key AI solutions with Ubuntu. Starting with Ubuntu 26.04 LTS, users can get [NVIDIA CUDA](https://canonical.com/blog/canonical-announces-it-will-support-and-distribute-nvidia-cuda-in-ubuntu), [AMD ROCm](https://canonical.com/blog/canonical-amd-rocm-ai-ml-hpc-libraries), and [NVIDIA DOCA-OFED](https://canonical.com/blog/canonical-announces-it-will-distribute-nvidia-doca-ofed-in-ubuntu) each with a single command.\n\n### GPGPU frameworks\n\nNVIDIA CUDA and AMD ROCm are frameworks for general-purpose computing on graphics processing units (GPGPU). They are the critical software layers that enable developers to harness the massive throughput of NVIDIA and AMD GPUs for AI workloads.\n\nHistorically, installing these frameworks required multi-step processes, and navigating dependency and compatibility issues could often prove challenging, especially for inexperienced users. But with Ubuntu 26.04 LTS, NVIDIA CUDA or AMD ROCm can each be installed with just one **apt install **command.\n\nThe new distribution model can save teams hours or even days on GPGPU framework setup, so organizations can start gaining value from GPUs faster. Canonical also ensures that users have smooth upgrade paths, so they can be confident when updating, and get the benefits of the latest features of these platforms.\n\nHave questions about AMD ROCm on Ubuntu? [We’ve just published a deep dive.](https://ubuntu.com/blog/amd-rocm-on-ubuntu)\n\n### High-performance networking\n\nFor organizations with large-scale [AI factories](https://canonical.com/knowledge/data-and-ai-ml/what-is-an-ai-factory) and HPC clusters, NVIDIA DOCA-OFED is among the go-to high-performance networking stacks. However, traditionally, the tradeoff for enabling ultra-low latency and high-throughput data transfers was the complexity of setup and maintenance. System administrators had to manage networking drivers through external installers or complex manual builds, potentially leading to version conflicts or kernel mismatch issues during OS updates.\n\nNow that NVIDIA DOCA-OFED can be installed seamlessly, the entire lifecycle management is simplified. Alongside rapid installation, the new workflow solves common operational pain points like kernel drift, driver incompatibility, and CI breakage following kernel or OS upgrades. Infrastructure teams can deliver speed and stability, while saving resources.\n\n## Optimized for hardware and humans\n\nJon Seager, Canonical’s VP of Engineering, has written recently about [the future of AI in Ubuntu](https://discourse.ubuntu.com/t/the-future-of-ai-in-ubuntu/81130). He signs off by stating that “Ubuntu is not becoming an AI product.” But what we are committed to is making Ubuntu an *enabler* for AI. Whether it’s at the silicon level with deep optimization for every architecture, or at the user level with streamlined toolkit adoption and lifecycle management, Ubuntu is the software layer that underpins an effective AI infrastructure strategy. It can help you get more tokens per watt, and beyond that, it can help you get to value faster and help bring down the operating costs for your stack.\n\nIf you’d like to learn more about AI infrastructure best practices, and how Ubuntu can fit into your AI strategy, [read the enterprise guide to private AI infrastructure](https://ubuntu.com/engage/open-source-ai-infrastructure).\n\n###\n[Enterprise AI, simplified](https://canonical.com/solutions/ai)\n\nAI doesn’t have to be difficult. Accelerate innovation with an end-to-end stack that delivers all the open source tooling you need for the entire AI/ML lifecycle.\n\n##### Newsletter signup\n\n### Related posts\n\n####\n[Let’s meet at AI4 and talk about AI infrastructure with open source](/blog/canonical-ubuntu-ai4-2025)\n\nDate: 11 – 13 August 2025 Booth: 353 Book a meeting You know the old saying: what happens in Vegas… transforms your AI journey with trusted open source. On...\n\n####\n[Securing AI agent workflows on Ubuntu with the new NVIDIA OpenShell snap](/blog/nvidia-openshell-ubuntu-announcement)\n\nBy packaging OpenShell as a snap, Canonical is enabling enterprises to confidently run next-generation agentic workflows across local devices, hybrid...\n\n####\n[Run NVIDIA Nemotron 3 Nano Omni locally in a single command](/blog/nvidia-nemotron-3-nano-omni)\n\nToday, NVIDIA introduced the NVIDIA Nemotron™ 3 Nano Omni, a highly-efficient multimodal model designed to understand and reason across video, audio, images,...", "url": "https://wpnews.pro/news/beyond-tokens-per-watt-using-ubuntu-26-04-lts-for-ai", "canonical_source": "https://ubuntu.com//blog/beyond-tokens-per-watt", "published_at": "2026-06-05 10:44:14+00:00", "updated_at": "2026-06-05 11:07:22.911131+00:00", "lang": "en", "topics": ["artificial-intelligence", "machine-learning", "ai-infrastructure", "ai-chips", "mlops"], "entities": ["Freyja Cooper", "Ubuntu", "Ubuntu 26.04 LTS"], "alternates": {"html": "https://wpnews.pro/news/beyond-tokens-per-watt-using-ubuntu-26-04-lts-for-ai", "markdown": "https://wpnews.pro/news/beyond-tokens-per-watt-using-ubuntu-26-04-lts-for-ai.md", "text": "https://wpnews.pro/news/beyond-tokens-per-watt-using-ubuntu-26-04-lts-for-ai.txt", "jsonld": "https://wpnews.pro/news/beyond-tokens-per-watt-using-ubuntu-26-04-lts-for-ai.jsonld"}}