{"id":"75276870-befb-4e12-bc67-e874346ea852","url":"https://api.tumn.com/news/75276870-befb-4e12-bc67-e874346ea852","title":"Meta Andromeda: Powering the Next Generation of Ads Retrieval Through Model-System Co-Design","summary":"This post describes the next generation of Meta Andromeda, Meta’s personalized ads retrieval engine, and how co-designing the model, system, and GPU hardware unlocks a new efficiency frontier for ads retrieval. In production, the system scores every eligible ad with near ranking-","sections":[],"faq":[],"sources":[{"url":"https://engineering.fb.com/2026/10/28/ml-applications/meta-andromeda-next-gen-ads-retrieval-gpu-co-design/","title":"Meta Engineering","publishedAt":"2026-10-28T16:00:01.000Z"}],"author":"TUMN 큐레이션","aiAssisted":false,"reviewStatus":"automated","reviewNote":"원문 제목과 공개된 발췌를 인용하고 원문으로 연결합니다. 내용을 재작성하거나 요약하지 않았습니다.","publishedAt":"2026-10-28T16:00:01.000Z","modifiedAt":"2026-10-28T20:26:31.000Z","checkedAt":"2026-10-28T20:26:31.810Z","language":"ko","kind":"link","sourceLabel":"AI 공식","profile":null,"via":null,"formats":{"json":"https://api.tumn.com/news/75276870-befb-4e12-bc67-e874346ea852.json","markdown":"https://api.tumn.com/news/75276870-befb-4e12-bc67-e874346ea852.md"}}