Migrating the Hugging Face Hub from Git LFS to Xet
Hugging Face is migrating its model and dataset hosting infrastructure from Git LFS to Xet, a content-addressed storage system designed for large binary files. The migration aims to improve upload/download speeds, deduplication, and storage efficiency for the large model weights and datasets hosted on the Hub. This represents a significant infrastructure change affecting how millions of AI artifacts are stored and accessed by the community.
Related guides (3)
Related events (8)
Xet Storage Integration on Hugging Face Hub
Hugging Face has integrated Xet, a chunk-based deduplication storage backend, into the Hub to improve large model file storage and transfer efficiency. The integration aims to reduce redundant data storage and speed up uploads/downloads for large model weights by splitting files into content-addressed chunks. This is an infrastructure improvement relevant to the open-weights ecosystem where multi-gigabyte model files are common.
XetHub Joins Hugging Face
XetHub, a company specializing in large-scale file storage and versioning for ML datasets and models, is being acquired by Hugging Face. The acquisition is intended to strengthen Hugging Face's infrastructure for hosting and managing large model and dataset repositories. This move reflects ongoing consolidation in the AI tooling and infrastructure space around the Hugging Face platform.
Introducing Storage Buckets on the Hugging Face Hub
Hugging Face is launching Storage Buckets, a new feature on the Hub that provides object storage capabilities for AI/ML workflows. This expands the Hub's infrastructure offerings beyond model and dataset repositories, enabling users to store arbitrary files and artifacts. The feature targets teams managing large-scale AI pipelines who need integrated storage alongside their models and datasets.
Improving Hugging Face Model Access for Kaggle Users
Hugging Face has announced an integration improvement that streamlines how Kaggle users access models from the Hugging Face Hub. The update appears to reduce friction for practitioners using Kaggle notebooks and compute environments to work with Hugging Face-hosted models. This represents a platform-level partnership move between two major ML community hubs.
Hugging Face redesigns hf CLI to be agent-optimized for Hub interactions
Hugging Face published a blog post describing design decisions behind making the hf CLI agent-friendly for interacting with the Hub. The post covers how the CLI is being structured to work well in agentic workflows where LLMs or automated systems issue commands programmatically. This is relevant to the growing ecosystem of AI agents that need to retrieve, upload, or manage models and datasets.
Announcing New Hugging Face and KerasHub Integration
Hugging Face and KerasHub have announced a new integration enabling users to access Hugging Face models and datasets directly through the Keras ecosystem. This partnership bridges two major ML frameworks, allowing Keras users to leverage the Hugging Face Hub's model repository without leaving the Keras workflow. The integration is aimed at reducing friction for practitioners who prefer Keras-based training and inference pipelines.
huggingface_hub v1.0: Five Years of Building the Foundation of Open Machine Learning
Hugging Face has released huggingface_hub v1.0, marking a major milestone for the Python client library that underpins access to the Hugging Face Hub ecosystem. The v1.0 designation signals API stability and maturity after five years of development. This library is a foundational piece of open-source ML infrastructure, enabling model downloads, dataset access, and repository management across the broader ML community.
Hugging Face and FriendliAI Partner to Supercharge Model Deployment on the Hub
Hugging Face and FriendliAI have announced a partnership to integrate FriendliAI's inference infrastructure directly into the Hugging Face Hub. The collaboration aims to simplify and accelerate model deployment for developers accessing models through the Hub. This expands the ecosystem of inference providers available on Hugging Face's platform.