{"type":"video","version":"1.0","html":"<iframe src=\"https://www.loom.com/embed/50891d4548bc4658b21d6cb8bdadfbfc\" frameborder=\"0\" width=\"1064\" height=\"798\" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>","height":798,"width":1064,"provider_name":"Loom","provider_url":"https://www.loom.com","thumbnail_height":798,"thumbnail_width":1064,"thumbnail_url":"https://cdn.loom.com/sessions/thumbnails/50891d4548bc4658b21d6cb8bdadfbfc-1688590534633.gif","duration":404.4333333333342,"title":"Everinfer Demo: Fast Serverless AI Inference 🚀","description":"Everinfer demo: 1:00 — 65ms cold start with the simplest model, 1:46 — ~1s cold start with BERT, 2:45 — 377ms cold start for CV model, 3:07 — batch processing for 2x speedup."}