According to Gartner*, 40% of generative AI (GenAI) solutions will be multimodal by 2027.¹ This session offers a closer look into running multimodal generative AI on the AI PC, a laptop built with a CPU, GPU, and NPU to handle AI tasks locally and more efficiently.

Specifically, this webinar focuses on using the GPU and NPU in combination with OpenVINO™ toolkit to develop, optimize, and deploy image-based GenAI models—such as Stable Diffusion* and latent consistency models—using multimodal learning.

Topics covered in the webinar include:

• Understand how AI acceleration technologies are integrated across AI PC hardware
• Parallels and comparisons between GenAI algorithms and visual, multimodal capabilities on AI PCs across text, audio, and images, and lessons learned during the development and optimization processes
• Practical knowledge in implementing and deploying state-of-the-art AI models using the OpenVINO toolkit

This webinar demonstrates reproducible source code that showcases the performance and power efficiency of AI applications on the AI PC, including multimodal applications such as Kosmos-2 and LLaVA.

Skill level: Intermediate