Meta’s Memory Layer pushes Instagram Reels item coverage to 100% and freshness to 20 seconds
_reachsumit · x · 2026-07-29
Meta’s Memory Layer trains an in-model key-value cache alongside the recommender itself, so training and serving share one source of truth for item embeddings.
- Deployed on Instagram Reels.
- Raises coverage from 96% to 100%.
- Cuts embedding freshness from about 5 minutes to about 20 seconds.
- Narrows the training-serving NE gap by up to 86% and more than doubles recall for the freshest content.
More from Infra
- NVIDIA launches RTX Spark for local AI on Windows PCs — ostrisai · 2026-10-16
- SkyRL ships Day-0 RL support for Beam with +62% LoRA decode throughput in vLLM — vllm_project · 2026-10-16
- DFlash2 delivers up to 3.03x decode-only speedup for Beam across five benchmarks — vllm_project · 2026-10-16
- vLLM's Beam serving kernels cut GB200 forward time from 8.15 to 6.47 ms at TP8 batch 128 — vllm_project · 2026-10-16
- Reflection's 501B MoE Beam hits 463 tok/s on 8x GB200 with NVFP4 at day 0 — zhyncs42 · 2026-10-16
- Gemma 4 31B QAT on one 24GB card: 46 tok/s plain, ~80 with draft model on RTX 4090 — Subject_Poet_3861 · 2026-10-16