TStore: Rethinking AI Model Hub with Tensor-Centric Compression

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Lan, Tingfeng, Wang, Zirui, Zheng, Yunjia, Su, Zhaoyuan, Yang, Juncheng, Cheng, Yue
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909038230372352
author Lan, Tingfeng
Wang, Zirui
Zheng, Yunjia
Su, Zhaoyuan
Yang, Juncheng
Cheng, Yue
author_facet Lan, Tingfeng
Wang, Zirui
Zheng, Yunjia
Su, Zhaoyuan
Yang, Juncheng
Cheng, Yue
contents Modern AI models are growing rapidly in size and redundancy, leading to significant storage and distribution challenges in model hubs. We present TStore, a tensor-centric system for reducing storage overhead through fine-grained deduplication and compression. TStore leverages tensor-level fingerprinting and clustering to identify redundancy across models without requiring annotations. Our design enables efficient storage reduction while preserving model usability and performance. Experiments on real-world model repositories demonstrate substantial storage savings with minimal overhead.
format Preprint
id arxiv_https___arxiv_org_abs_2604_17104
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle TStore: Rethinking AI Model Hub with Tensor-Centric Compression
Lan, Tingfeng
Wang, Zirui
Zheng, Yunjia
Su, Zhaoyuan
Yang, Juncheng
Cheng, Yue
Distributed, Parallel, and Cluster Computing
Artificial Intelligence
Machine Learning
D.4.2; E.5; I.2.6
Modern AI models are growing rapidly in size and redundancy, leading to significant storage and distribution challenges in model hubs. We present TStore, a tensor-centric system for reducing storage overhead through fine-grained deduplication and compression. TStore leverages tensor-level fingerprinting and clustering to identify redundancy across models without requiring annotations. Our design enables efficient storage reduction while preserving model usability and performance. Experiments on real-world model repositories demonstrate substantial storage savings with minimal overhead.
title TStore: Rethinking AI Model Hub with Tensor-Centric Compression
topic Distributed, Parallel, and Cluster Computing
Artificial Intelligence
Machine Learning
D.4.2; E.5; I.2.6
url https://arxiv.org/abs/2604.17104