---
slug: /tir/object-storage
sidebar_position: 6
title: "Object Storage"
description: >-
  S3-compatible EOS buckets for AI/ML data, with versioning, object lock,
  lifecycle rules, and presigned URLs. Previously known as Datasets (EOS).
---

import { ObjectStorageBadges, ObjectStorageQuickStart, ObjectStorageCapabilities, ObjectStorageAPIDoc, ObjectStorageBilling } from './ObjectStorageCards'

# Object Storage

Object Storage, previously known as **Datasets (EOS)**, is backed by a bucket in E2E Object Storage (EOS), an S3-compatible store. Use it for training data, checkpoints, and artifacts that need to grow without a fixed size, be read by several workloads at once, or be reached from outside TIR with an S3 CLI or SDK.

In the TIR portal it lives under **Storage > Object Storage**. Each bucket has a **bucket name**, which is what EOS knows it as, and an **alias**, its name inside TIR.

<ObjectStorageBadges />

<ObjectStorageQuickStart />

---

## Where it is used

Buckets always mount **read-only**. Where the bucket appears depends on what you attach it to:

- **[Instances](/docs/tir/Nodes/instances) (containers) and [Virtual Machines](/docs/tir/VirtualMachines/virtual-machines):** always at `/datasets/<alias>`.
- **[Slurm clusters](/docs/tir/SlurmCluster/):** at a mount path you choose when you attach the bucket, such as `/data`.

Buckets are also the destination for [Data Syncer](/docs/tir/DataSyncer/) transfers and the store behind fine-tuning data.

---

## API Reference

<ObjectStorageAPIDoc />

## Billing & Plans

<ObjectStorageBilling />

<ObjectStorageCapabilities />