Nvidia recently announced a collaboration with Microsoft to power personalized AI applications on Windows through Copilot. This collaboration is not limited to just Nvidia, as other GPU vendors such as AMD and Intel will also benefit from this partnership. The Windows Copilot Runtime will now support GPU acceleration, making it easier for GPUs to apply their AI capabilities to apps on the operating system. This collaboration aims to provide application developers with easy API access to GPU-accelerated small language models (SLMs) that enable retrieval-augmented generation (RAG) capabilities powered by Windows Copilot Runtime.
In simpler terms, developers will now be able to utilize an API to leverage GPU acceleration for heavily personalized AI tasks on Windows. This includes tasks such as content summaries, automation, and generative AI. Nvidia has already introduced an RAG application called Chat with RTX that runs on its own graphics cards. With the support of Copilot Runtime, developers can explore more applications like Project G-Assist. Additionally, Nvidia has launched the RTX AI Toolkit, which is a suite of tools and SDKs for model customization. This collaboration opens up new possibilities for application developers to create more powerful and efficient AI applications.
The current landscape of client AI inference is dominated by Intel, AMD, and Qualcomm in laptops. However, GPUs have proven to be incredibly powerful for AI applications. Developers have the flexibility to choose where they want to deploy their AI applications – CPU, NPU, or GPU. With easier API access to GPU acceleration through Copilot Runtime, developers can harness the full potential of these components and create more advanced applications. This move not only benefits Nvidia but also other GPU vendors looking to capitalize on the growing demand for AI-powered applications.
It’s not just Nvidia that stands to gain from this collaboration. Other GPU vendors will also have access to GPU acceleration through Copilot Runtime. This means that AI capabilities will be accelerated not only by Nvidia RTX GPUs but also by AI accelerators from other hardware vendors. This opens up a world of fast and responsive AI experiences across different devices within the Windows ecosystem. While Microsoft currently requires 45 TOPs of NPU processing for entry into its AI-ready computer club, known as Copilot+, GPUs are not yet included in the requirement despite offering superior performance. However, with speculations about Nvidia potentially developing its ARM-based SoC, it is likely that Windows on ARM will leverage Nvidia’s integrated GPUs for its Copilot AI operations.
A preview API for GPU acceleration on Copilot Runtime is expected to be available later this year in a Windows developer build. This will allow developers to test and optimize their AI applications for GPU acceleration, paving the way for more efficient and high-performance AI experiences on Windows devices.Overall, the collaboration between Nvidia and Microsoft to power personalized AI applications on Windows through Copilot is a significant step towards unlocking the full potential of GPU acceleration for AI tasks. With easier API access and support for other GPU vendors, developers can expect a new era of innovative and powerful AI applications on the Windows platform.
Leave a Reply