1、1 February 2025 AI disruption is driving innovation in on-device inference How the proliferation and evolution of generative models will transform the AI landscape and unlock value.Snapdragon and Qualcomm branded products are products of Qualcomm Technologies,Inc.and/or its subsidiaries.2 Contents E

2、xecutive summary.3 Quality AI models are now abundant and aordable.4 Innovations boost model quality and reduce development time and cost.4 Small models achieve big capabilities at the edge.5 The era of AI inference innovation is here.7 Qualcomm is set to be a leader in the AI inference era.8 Expand

3、ing across all key edge segments.9 Mobile.9 PCs.10 Automotive.10 Industrial IoT.11 Networking.11 Conclusion.11 3 Executive summary The introduction of DeepSeek R1,a cutting-edge reasoning AI model,has caused ripples throughout the tech industry.Thats because its performance is on par with or better

4、than state-of-the-art alternatives,disrupting the conventional wisdom around AI development.This pivotal moment is part of a broader trend that underscores the innovation in creating high-quality small language and multimodal reasoning models,and how theyre preparing AI for commercial applications a

5、nd on-device inference.The fact that these new models can run on devices accelerates scale and creates demand for powerful chips at the edge.Driving this shift are four major trends that are leading to a dramatic improvement in the quality,performance,and eiciency of AI models that can now run on de

6、vice:Todays state-of-the-art smaller AI models have superior performance.New techniques like model distillation and novel AI network architectures simplify the development process without sacrificing quality,allowing new models to outperform larger ones from a year ago,which could only operate on th