---
format: "aidr-story-markdown/v1"
id: "dba183e7e6624faa79adbf3eb13681be485aa2ccef176f6746ca1af76d635b02"
canonical_url: "https://aidr.today/dba183e7?lang=en"
title: "Meta FAIR Releases e-means: An Algorithm Beating k-means in 10x Fewer Epochs"
lang: "en"
requested_lang: "en"
available_langs: ["en","vi"]
translation_fallback: null
fallback_fields: []
published_at: "2026-11-10T23:20:19.000Z"
category: "Frameworks"
topics: ["meta","open-source","framework"]
source_urls: ["https://www.marktechpost.com/2026/11/10/meta-fair-e-means-k-means-faiss/"]
summary: "Meta FAIR released e-means, a mini-batch k-means replacement included in FAISS 1.16.0, which reaches Lloyd’s 25-epoch loss in 1.4 to 2.6 epochs with 10-45x GPU speedups and no extra hyperparameters."
---

# Meta FAIR Releases e\-means: An Algorithm Beating k\-means in 10x Fewer Epochs

> [Open the canonical story](<https://aidr.today/dba183e7?lang=en>)

**Published:** 2026-11-10T23:20:19.000Z
**Category:** Frameworks
**Topics:** meta, open\-source, framework

## Summary

Meta FAIR released e\-means, a mini\-batch k\-means replacement included in FAISS 1\.16\.0, which reaches Lloyd’s 25\-epoch loss in 1\.4 to 2\.6 epochs with 10\-45x GPU speedups and no extra hyperparameters\.

## Sources

- [Story source](<https://www.marktechpost.com/2026/11/10/meta-fair-e-means-k-means-faiss/>)