Unleash the superpowers of the AI PC
summary
Demonstrations run LLMs and agents locally using OpenVINO and DirectML with CPU, integrated GPU and NPU collaborating: Whisper speech runs on NPU, Phi-3 inference on iGPU, vector search on CPU. A ReAct agent calls RAG (150 BBC Sport articles indexed) and Gmail tools to reason and act, correcting a Champions League answer from 32 to 36 teams. Models use four‑bit quantization and speculative decoding; open‑source Python notebooks reproduce the demos.
AI-written summary · about
chapters (6)
- 0:00Introduction to AI PC
- 2:30Demo overview
- 3:38ReAct agent demonstration
- 6:42Retrieval augmented generation
- 9:35Multimodal and code insights
- 11:37Real world AI pillars
related talks
30mCloud? No Thanks! I’m Gonna Run GenAI on My AI PC
30mLook ma, no hands! Multimodal AI-Agents in the browser
17mAI/ML demo derby
01h 03mIt's All About AI
46mFullstack AI: The Building Blocks for Production Ready AI-Powered Software
14mReliable Agentic AI, the Open Way
02h 08mIntroduction to AI Agents
21mDemocratizing AI with Open Source Multi-Agent Systems
