DeepSeek’s R1 Model Now Optimized for NPU Running on Copilot+ Windows 11 PCs, Advanced Variants Coming Soon

DeepSeek’s R1 Model Now Optimized for NPU Running on Copilot+ Windows 11 PCs, Advanced Variants Coming Soon

DeepSeek’s R1 model has taken center stage in an investigation led by Microsoft, probing whether this AI model utilized data outputs from OpenAI. Despite this scrutiny, Microsoft is proceeding with the introduction of an NPU-optimized variant of the model for its Copilot+ PCs. Recently, the tech giant announced that DeepSeek-R1-Distill-Qwen-1.5B will be accessible through the Microsoft AI Toolkit, specifically designed for developers. While Microsoft plans to release more advanced models in the future, it is currently making measured advancements.

How to Successfully Run DeepSeek R1 on Windows 11

In a landscape where many U. S.companies are expressing concern over DeepSeek’s climbing popularity, Microsoft has strategically embraced the R1 AI model. The company detailed in its blog that this optimized version will be compatible with PCs powered by Qualcomm’s Snapdragon X chipsets, with plans to extend support to Intel’s Core Ultra 200V and additional processors. Following the release of DeepSeek-R1-Distill-Qwen-1.5B, Microsoft is set to roll out variants with 7 billion and 14 billion parameters, allowing developers to harness their full capabilities. To facilitate the implementation of DeepSeek on systems, Microsoft has outlined the following instructions:

“To see DeepSeek in action on your Copilot+ PC, simply download the AI Toolkit VS Code extension. The DeepSeek model optimized in the ONNX QDQ format will soon be available in AI Toolkit’s model catalog, pulled directly from Azure AI Foundry. You can download it locally by clicking the “Download” button. Once downloaded, experimenting with the model is as simple as opening the Playground, loading the “deepseek_r1_1_5″ model, and sending it prompts.”

Furthermore, Microsoft has implemented various enhancements to ensure that DeepSeek’s R1 efficiently runs locally on NPU-based hardware. For those interested in exploring Microsoft’s developments further, a source link is provided below. As for the ongoing investigation regarding DeepSeek’s potential training on OpenAI data, no new updates have been released, but we will keep you informed as the situation evolves, so stay tuned for future announcements.

For more details, visit the official Microsoft blog.

Source & Images

Leave a Reply

Your email address will not be published. Required fields are marked *