{"slug": "gem-training-how-meta-doubled-the-efficiency-of-its-llm-scale-ads-foundation", "title": "GEM Training: How Meta Doubled the Efficiency of Its LLM-Scale Ads Foundation Model", "summary": "Meta's Generative Ads Recommendation Model (GEM), the foundation model behind ads recommendations on Instagram and Facebook, now trains at LLM scale on several thousand of the latest-generation GPUs, achieving a doubling of end-to-end training efficiency to 20–25% Model FLOPs Utilization (MFU) while scaling training FLOPs 4x, as detailed in a post on Engineering at Meta.", "body_md": "Meta’s Generative Ads Recommendation Model (GEM), the foundation model behind ads recommendations across Instagram and Facebook, now trains at LLM scale on several thousand of the latest-generation GPUs. This post goes into the details on how we achieved: doubling end-to-end (E2E) training efficiency to 20–25% Model FLOPs Utilization (MFU) while scaling training FLOPs 4x in [...]\n\nThe post [GEM Training: How Meta Doubled the Efficiency of Its LLM-Scale Ads Foundation Model](https://engineering.fb.com/2026/08/03/ml-applications/training-gem-at-llm-scale-meta-ads-recommendation-foundation-model/) appeared first on [Engineering at Meta](https://engineering.fb.com).", "url": "https://wpnews.pro/news/gem-training-how-meta-doubled-the-efficiency-of-its-llm-scale-ads-foundation", "canonical_source": "https://engineering.fb.com/2026/08/03/ml-applications/training-gem-at-llm-scale-meta-ads-recommendation-foundation-model/", "published_at": "2026-08-03 18:00:17+00:00", "updated_at": "2026-08-24 23:14:37.583174+00:00", "lang": "en", "topics": ["machine-learning", "large-language-models", "ai-infrastructure", "ai-research"], "entities": ["Meta", "Generative Ads Recommendation Model (GEM)", "Instagram", "Facebook", "Engineering at Meta"], "alternates": {"html": "https://wpnews.pro/news/gem-training-how-meta-doubled-the-efficiency-of-its-llm-scale-ads-foundation", "markdown": "https://wpnews.pro/news/gem-training-how-meta-doubled-the-efficiency-of-its-llm-scale-ads-foundation.md", "text": "https://wpnews.pro/news/gem-training-how-meta-doubled-the-efficiency-of-its-llm-scale-ads-foundation.txt", "jsonld": "https://wpnews.pro/news/gem-training-how-meta-doubled-the-efficiency-of-its-llm-scale-ads-foundation.jsonld"}}