{"type":"link","version":"1.0","title":"Spare capacity should be spent speculating ahead; saturation should be spent scheduling well","author_name":"AI Archs","author_url":"https://ai-arch.pages.dev","provider_name":"AI Archs","provider_url":"https://ai-arch.pages.dev","url":"https://ai-arch.pages.dev/n/the-best-inference-optimization-depends-on-the-load-regime","thumbnail_url":"https://ai-arch.pages.dev/og/the-best-inference-optimization-depends-on-the-load-regime.png","thumbnail_width":1200,"thumbnail_height":630}