Hugging Face 基礎設施大升級：將 Hub 從 Git LFS 遷移至 Xet
Original: Migrating the Hub from Git LFS to Xet
The official Hugging Face blog has announced a major underlying technology migration: moving the large-file storage mechanism of the…
Hugging Face 宣布將其平台（Hub）的底層大檔案版本控制系統從 Git LFS 遷移至 Xet 技術。Git LFS 在面對數百 GB 的模型與超大數據集時常遇到效能瓶頸，而 Xet 透過「內容定義分塊（CDC）」技術實現區塊級去重，能顯著減少儲存空間並加速上傳與下載。此遷移將為開發者帶來更流暢的超大型 AI 資產管理體驗。
The official Hugging Face blog has announced a major underlying technology migration: moving the large-file storage mechanism of the Hugging Face Hub from the long-established Git LFS (Large File Storage) to Xet technology. This change aims to resolve the performance bottlenecks that modern AI's super-large models and massive datasets face in version control and data transfer.
Free shows the 3-line summary; Pro unlocks the full deep summary (~300 words) so you never have to click through.
See Pro plans →Want the original English / full article?
Read on Hugging Face Blog →Summaries are AI-generated; the original article is authoritative.