[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"tag-fp8-kv-cache":3},{"tag":4,"articles":10},{"id":5,"name":6,"slug":7,"article_count":8,"description_zh":9,"description_en":9},"4e572372-b2e7-4eb1-87c7-410f61778dd5","FP8 KV-cache","fp8-kv-cache",0,null,[11],{"id":12,"slug":13,"title":14,"summary":15,"category":16,"image_url":17,"cover_image":17,"language":18,"created_at":19},"381fb6c6-6da7-4444-831f-8c5eed8d685c","turboquant-vllm-comparison-fp8-kv-cache-zh","TurboQuant 與 FP8 實測結果","vLLM 首次大規模比較 TurboQuant 與 FP8 KV-cache。結果很直白：FP8 在速度上更穩，TurboQuant 的高壓縮版本則常掉準確率。","research","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1778839867551-4v9g.png","zh","2026-05-15T10:10:36.034569+00:00"]