[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"tag-on-device-ai":3},{"tag":4,"articles":11},{"id":5,"name":6,"slug":7,"article_count":8,"description_zh":9,"description_en":10},"dcc331c5-33f1-4f0c-95a3-19564116eb8f","on-device AI","on-device-ai",3,"裝置端 AI 指把推論放在手機、筆電、穿戴或邊緣裝置上執行，減少雲端依賴、延遲與資料外送。它牽涉小型語言模型、NPU、隱私保護與離線體驗，也正改變 Siri、行動助理與終端應用的設計方式。","On-device AI runs inference locally on phones, laptops, wearables, or edge devices instead of sending data to the cloud. It matters for latency, privacy, offline use, and hardware design, with small language models, NPUs, and assistants like Siri driving the shift.",[12,21],{"id":13,"slug":14,"title":15,"summary":16,"category":17,"image_url":18,"cover_image":18,"language":19,"created_at":20},"7176e23d-dcfc-473b-adb0-8e8124535077","apple-50-how-it-can-still-win-in-ai-en","Apple at 50: How it can still win in AI","Apple is betting on on-device AI and a Gemini-powered Siri reboot after years of delay. Former insiders think that may still be enough.","industry","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1775630405753-7gcx.png","en","2026-04-08T06:39:44.061962+00:00",{"id":22,"slug":23,"title":24,"summary":25,"category":26,"image_url":27,"cover_image":27,"language":19,"created_at":28},"895c150c-569e-4fdf-939d-dade785c990e","small-language-models-transform-ai-en","Small Language Models: Llama 3.2 and Phi-3 Transform AI","Llama 3.2 and Phi-3 redefine AI, shifting focus from cloud-heavy solutions to efficient, privacy-focused on-device applications.","model-release","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1774497782114-y6ap.png","2026-03-25T16:30:26.688313+00:00"]