Articles

S3-Compatible Cloud Object Storage Providers (Hyperscalers and Regional Clouds)

Jun 28, 2026

The large platform clouds all run object storage, and most expose an Amazon S3-compatible endpoint alongside their native API. That means you can point an S3 client at Alibaba, Tencent, Huawei, IBM, Oracle, Google, Yandex, and many regional clouds by changing the endpoint URL and keys. This page lists the cloud S3-compatible services, their endpoint formats, and how to connect each one.

This is one category in our complete list of S3-compatible storage providers. For independent object-storage clouds, hosting platforms, self-hosted servers, enterprise appliances, and decentralized storage, see the other category pages linked at the end.

These services are intended to connect through Blober's generic S3-Compatible connector when they support the common S3 operations Blober uses. Paste the endpoint and keys, then browse to confirm. Confirm the exact endpoint in each provider's console, since regions and hostnames change over time. The endpoint formats below come from each provider's own documentation, cross-checked against current S3 client references^[1].

Each entry notes an addressing style. Virtual-hosted puts the bucket in the subdomain (https://my-bucket.s3.example.com); path-style puts it in the URL path (https://s3.example.com/my-bucket). Almost every cloud here uses virtual-hosted. The endpoint formats below show hostnames; in Blober, include https:// in the endpoint field. The endpoint setup notes explain how Blober picks the right one.

Alibaba Cloud OSS (Object Storage Service)

Alibaba Cloud Object Storage Service, formerly Aliyun OSS, is the dominant object store in China and across Alibaba's global regions.

Endpoint format: oss-<region>.aliyuncs.com (for example oss-cn-hangzhou.aliyuncs.com, oss-us-east-1.aliyuncs.com, oss-eu-central-1.aliyuncs.com).
Regions: dozens across mainland China, plus Singapore, Tokyo, Frankfurt, London, Dubai, Silicon Valley, and Virginia.
Addressing: virtual-hosted (bucket in subdomain).
Notes: OSS offers Standard, Infrequent Access, Archive, and Cold Archive tiers. Use the matching regional endpoint for your bucket^[2].

Tencent Cloud COS (Cloud Object Storage)

Tencent Cloud Object Storage is the object store behind much of China's app ecosystem, with a global region footprint.

Endpoint format: cos.<region>.myqcloud.com (for example cos.ap-guangzhou.myqcloud.com, cos.ap-singapore.myqcloud.com, cos.na-ashburn.myqcloud.com).
Regions: Beijing, Shanghai, Guangzhou, Chengdu, Hong Kong, Singapore, Mumbai, Seoul, Bangkok, Tokyo, Silicon Valley, Virginia, Toronto, Frankfurt.
Addressing: virtual-hosted.
Notes: there is also a Global Acceleration endpoint (cos.accelerate.myqcloud.com). Standard, Infrequent Access, and Archive tiers are available^[3].

Huawei Cloud OBS (Object Storage Service)

Huawei Cloud OBS is a high-capacity object store with an S3-compatible interface, used widely across Asia, Latin America, and EMEA.

Endpoint format: obs.<region>.myhuaweicloud.com (for example obs.af-south-1.myhuaweicloud.com, obs.ap-southeast-3.myhuaweicloud.com).
Regions: Johannesburg, Bangkok, Singapore, Hong Kong, Shanghai, Beijing, Guangzhou, plus Latin America (Sao Paulo, Mexico City, Santiago, Buenos Aires, Lima) and Moscow.
Addressing: virtual-hosted.
Notes: OBS underpins several rebranded regional clouds (Open Telekom Cloud and SberCloud, listed below), so the connection pattern is similar^[4].

IBM Cloud Object Storage (COS)

IBM Cloud Object Storage, descended from Cleversafe, disperses data across locations and is accessed through an S3-compatible API. It is also sold as on-premises software.

Endpoint format: s3.<region>.cloud-object-storage.appdomain.cloud (for example s3.us-east.cloud-object-storage.appdomain.cloud, s3.eu-de.cloud-object-storage.appdomain.cloud).
Regions: US East and South, EU (Frankfurt, Amsterdam, Milan), Great Britain, plus single-site endpoints in many cities. Private endpoints exist for in-cloud traffic.
Addressing: virtual-hosted.
Notes: IBM COS supports IAM API-key authentication in addition to HMAC access keys. Storage classes include Standard, Vault, Cold Vault, and Smart Tier^[5].

Oracle Cloud Infrastructure (OCI) Object Storage

Oracle Cloud Infrastructure Object Storage offers an Amazon S3 Compatibility API in front of its native object store.

Endpoint format: <namespace>.compat.objectstorage.<region>.oci.customer-oci.com for path-style, or <bucket>.vhcompat.objectstorage.<region>.oci.customer-oci.com for virtual-hosted. The <namespace> is your tenancy's unique Object Storage namespace.
Regions: Ashburn, Phoenix, Frankfurt, London, Tokyo, and the rest of OCI's global regions.
Addressing: both path and virtual-hosted are supported.
Notes: authenticate with a Customer Secret Key (an access key and secret pair). If your tool cannot set the region identifier, set it to us-east-1 or leave it blank to use your home region. SigV2 is not supported^[6].

Google Cloud Storage (XML / S3 Interoperability)

Google Cloud Storage is not S3 by default, but its XML API has an S3-compatible interoperability mode that works with HMAC keys.

Endpoint format: storage.googleapis.com.
Regions: global; the endpoint is single, and the bucket's location is set when you create it.
Addressing: virtual-hosted.
Notes: create an HMAC key for a service account to get an access key and secret. Object versioning listing and a few advanced behaviors differ from AWS, so treat GCS as S3-compatible for the common operations (browse, upload, download, copy) rather than a perfect clone^[7].

Yandex Object Storage

Yandex Cloud's object storage is S3-compatible and popular across Russia and the CIS.

Endpoint format: storage.yandexcloud.net.
Regions: ru-central1 and its availability zones; the endpoint is single.
Addressing: virtual-hosted.
Notes: Standard (hot) and Cold storage classes are available. Use static access keys generated for a service account^[8].

VK Cloud Object Storage (formerly Mail.ru Cloud Solutions)

VK Cloud, previously Mail.ru Cloud Solutions, runs Hotbox and Icebox object storage tiers behind an S3-compatible endpoint.

Endpoint format: hb.bizmrg.com (Hotbox) and ib.bizmrg.com (Icebox), with newer hb.<region>.vkcs.cloud hosts.
Regions: Moscow.
Addressing: virtual-hosted.
Notes: Hotbox is the hot tier, Icebox the cold tier. Confirm the current endpoint in the VK Cloud console.

Baidu AI Cloud BOS (Baidu Object Storage)

Baidu Object Storage provides an S3-compatible interface for one of China's major clouds.

Endpoint format: s3.<region>.bcebos.com (for example s3.bj.bcebos.com for Beijing, s3.gz.bcebos.com for Guangzhou).
Regions: Beijing, Baoding, Suzhou, Guangzhou, Hong Kong, and more.
Addressing: virtual-hosted.
Notes: Standard, Infrequent Access, Cold, and Archive tiers are available.

JD Cloud Object Storage (OSS)

JD Cloud, the cloud arm of JD.com, exposes an S3-compatible object storage service.

Endpoint format: s3.<region>.jdcloud-oss.com (for example s3.cn-north-1.jdcloud-oss.com).
Regions: North China, East China, and South China.
Addressing: virtual-hosted.
Notes: confirm the regional endpoint in the JD Cloud console.

China Mobile Ecloud EOS (Elastic Object Storage)

China Mobile's Ecloud Elastic Object Storage is a large, region-dense S3-compatible service across mainland China.

Endpoint format: eos-<city>-1.cmecloud.cn (for example eos-wuxi-1.cmecloud.cn, eos-beijing-1.cmecloud.cn).
Regions: dozens, from Suzhou and Shanghai to Chengdu, Guiyang, and Xian.
Addressing: virtual-hosted.
Notes: the location constraint must match the endpoint when creating buckets^[9].

Kingsoft Cloud KS3

Kingsoft Cloud Standard Storage Service (KS3) is an S3-compatible object store widely used in China.

Endpoint format: ks3-<region>.ksyuncloud.com or ks3-<region>.ksyun.com (for example ks3-cn-beijing.ksyuncloud.com).
Regions: Beijing, Shanghai, Guangzhou, Hong Kong, and international points.
Addressing: virtual-hosted.
Notes: confirm the current endpoint host in the Kingsoft console.

China Telecom CTYun OOS

China Telecom's CTYun (Tianyi Cloud) Object-Oriented Storage is an S3-compatible service across China Telecom's regions.

Endpoint format: oos-<region>.ctyunapi.cn (for example oos-sccd.ctyunapi.cn).
Regions: many across mainland China.
Addressing: virtual-hosted.
Notes: confirm the regional endpoint in the CTYun console.

NAVER Cloud Object Storage

NAVER Cloud Platform, South Korea's major cloud, runs an S3-compatible object storage service.

Endpoint format: <region>.object.ncloudstorage.com (for example kr.object.ncloudstorage.com).
Regions: Korea, plus other NAVER Cloud regions.
Addressing: virtual-hosted.
Notes: generate API keys in the NAVER Cloud console.

Open Telekom Cloud OBS

Open Telekom Cloud, operated by T-Systems on Huawei technology, exposes Huawei OBS under its own endpoints.

Endpoint format: obs.<region>.otc.t-systems.com (for example obs.eu-de.otc.t-systems.com).
Regions: Germany (Biere, Magdeburg), the Netherlands, and Switzerland.
Addressing: virtual-hosted.
Notes: the OBS lineage means it behaves like Huawei OBS, with European data residency.

SberCloud OBS

SberCloud Advanced (Cloud.ru) runs object storage on Huawei OBS technology with Russian endpoints.

Endpoint format: obs.<region>.hc.sbercloud.ru (for example obs.ru-moscow-1.hc.sbercloud.ru).
Regions: Moscow.
Addressing: virtual-hosted.
Notes: behaves like Huawei OBS with Russian data residency.

Netease NOS (Netease Object Storage)

Netease Object Storage offers an S3-compatible interface used by Netease's services and third parties.

Endpoint format: regional NOS endpoints (for example nos-eastchina1.126.net).
Regions: several across China.
Addressing: virtual-hosted (path-style addressing is disabled by the provider).
Notes: confirm the exact endpoint in the Netease console.

A Note on Microsoft Azure Blob Storage

Azure Blob Storage is the obvious omission here, and it is deliberate. Azure Blob uses its own API, not the S3 API, so it is not S3-compatible in the drop-in sense. The concepts line up (a container maps to a bucket, a blob to an object), but an S3 client cannot talk to Azure Blob without a translation gateway. If you need to move data between S3 and Azure, Blober bridges the two directly. See How to Transfer AWS S3 to Azure Blob.

Frequently Asked Questions

Is Google Cloud Storage S3-compatible? Through its XML API interoperability mode, yes, for the common operations. You create an HMAC key and point your S3 client at storage.googleapis.com. A few advanced behaviors differ from AWS S3.

Is Alibaba OSS the same as AWS S3? OSS speaks the S3 API, so S3 tools work against it by changing the endpoint and keys. It is run by Alibaba, not Amazon, and its tier names and some features are its own.

Why do Open Telekom Cloud and SberCloud look like Huawei OBS? Because they are built on Huawei OBS technology and rebranded for their regions. The S3-compatible connection pattern is the same, with different endpoints and data residency.

Can Blober move data from one cloud's object storage to another? Yes. Point Blober at a source endpoint and a destination endpoint and it copies between them directly for normal object transfers. Provider-specific S3 feature gaps can still matter for advanced workflows.

Get Blober

Connect to any cloud's S3-compatible object storage by URL and move data between providers directly, without filling your local disk.

Download Blober at blober.io

S3-Compatible Object Storage from Web Hosts and VPS Providers

Jun 28, 2026

If you already rent servers from a hosting or VPS company, there is a good chance they also sell S3-compatible object storage right next to your droplets, instances, or dedicated boxes. Keeping storage with your compute reduces latency and often avoids cross-provider egress. This page lists the hosting and VPS object-storage services, their endpoint formats, and how to connect each one.

This is one category in our complete list of S3-compatible storage providers. DigitalOcean Spaces has a preconfigured connector in Blober; the rest use the generic S3-Compatible connector. Confirm endpoints in each provider's console, since regions change over time. The endpoint formats below come from each provider's own documentation, cross-checked against current S3 client references^[1].

Each entry notes an addressing style. Virtual-hosted puts the bucket in the subdomain (https://my-bucket.s3.example.com); path-style puts it in the URL path (https://s3.example.com/my-bucket). Almost every host here uses virtual-hosted. The endpoint formats below show hostnames; in Blober, include https:// in the endpoint field. The endpoint setup notes explain how Blober picks the right one.

DigitalOcean Spaces

DigitalOcean Spaces is S3-compatible object storage with a built-in CDN, sold alongside DigitalOcean droplets.

Endpoint format: <region>.digitaloceanspaces.com (for example nyc3.digitaloceanspaces.com, fra1.digitaloceanspaces.com).
Regions: NYC3, SFO2, SFO3, AMS3, FRA1, SGP1, SYD1, BLR1, ATL1, LON1, TOR1, and more.
Addressing: virtual-hosted.
Notes: Blober has a preconfigured DigitalOcean Spaces connector. Spaces has no built-in cross-region replication; see DigitalOcean Spaces Regions, Replication, and Backup^[2].

Linode / Akamai Object Storage

Linode (now part of Akamai) offers S3-compatible object storage across many global regions.

Endpoint format: <region>.linodeobjects.com (for example us-east-1.linodeobjects.com, eu-central-1.linodeobjects.com).
Regions: Newark, Atlanta, Chicago, Los Angeles, Miami, Seattle, Washington DC, Amsterdam, Frankfurt, London, Madrid, Milan, Paris, Stockholm, Chennai, Jakarta, Osaka, Singapore, Sao Paulo, Melbourne.
Addressing: virtual-hosted.
Notes: one of the widest region footprints among VPS providers^[3].

Vultr Object Storage

Vultr Object Storage is S3-compatible storage with a CDN, billed with generous included transfer.

Endpoint format: <region>.vultrobjects.com (for example ewr1.vultrobjects.com, ams1.vultrobjects.com).
Regions: New Jersey, Silicon Valley, Amsterdam, Singapore, Bangalore, Delhi, and more.
Addressing: virtual-hosted.
Notes: includes an archival storage tier with lifecycle policies. Confirm the regional host in the Vultr console^[4].

OVHcloud Object Storage

OVHcloud offers S3-compatible object storage across its European, North American, and Asia-Pacific regions.

Endpoint format: s3.<region>.io.cloud.ovh.net (for example s3.gra.io.cloud.ovh.net for Gravelines, s3.de.io.cloud.ovh.net for Frankfurt). US regions use io.cloud.ovh.us.
Regions: Gravelines, Roubaix, Strasbourg, Paris, Frankfurt, London, Warsaw, Beauharnois, Toronto, Singapore, Sydney, Mumbai, plus US (Virginia, Oregon) and a cold archive region.
Addressing: virtual-hosted.
Notes: storage classes include Standard, High Performance, Standard Infrequent Access, and archive tiers^[5].

Hetzner Object Storage

Hetzner Object Storage is low-cost S3-compatible storage in Hetzner's European data centers.

Endpoint format: <region>.your-objectstorage.com (for example hel1.your-objectstorage.com, fsn1.your-objectstorage.com, nbg1.your-objectstorage.com).
Regions: Helsinki, Falkenstein, Nuremberg.
Addressing: virtual-hosted.
Notes: a popular budget option for European workloads^[6].

Scaleway Object Storage

Scaleway Object Storage is S3-compatible storage from the French cloud provider, with a Glacier cold tier.

Endpoint format: s3.<region>.scw.cloud (for example s3.nl-ams.scw.cloud, s3.fr-par.scw.cloud, s3.pl-waw.scw.cloud).
Regions: Amsterdam, Paris, Warsaw.
Addressing: virtual-hosted.
Notes: the GLACIER storage class uploads directly to Scaleway's cold tier; restore before reading^[7].

Contabo Object Storage

Contabo Object Storage is budget S3-compatible storage (built on Ceph) with included transfer.

Endpoint format: <region>.contabostorage.com (for example eu2.contabostorage.com, usc1.contabostorage.com, sin1.contabostorage.com).
Regions: Europe, United States, and Asia.
Addressing: virtual-hosted.
Notes: unlimited inbound and generous outbound transfer are included in the flat monthly rate. Confirm your exact endpoint in the Contabo console^[8].

Exoscale SOS (Simple Object Storage)

Exoscale Simple Object Storage is S3-compatible storage from the Swiss cloud provider, with European zones.

Endpoint format: sos-<zone>.exo.io (for example sos-ch-gva-2.exo.io, sos-de-fra-1.exo.io).
Regions: Geneva, Zurich, Frankfurt, Munich, Vienna, Sofia.
Addressing: virtual-hosted.
Notes: strong fit for Swiss and EU data residency^[9].

UpCloud Object Storage

UpCloud Object Storage is S3-compatible storage from the Finnish cloud provider.

Endpoint format: <region>.upcloudobjects.com (the exact host is shown in your console).
Regions: several across Europe, the US, and Asia-Pacific.
Addressing: virtual-hosted.
Notes: confirm the per-region endpoint host in the UpCloud control panel.

IONOS Cloud S3 Object Storage

IONOS S3 Object Storage is S3-compatible storage from the German provider IONOS.

Endpoint format: s3-<region>.ionoscloud.com (for example s3-eu-central-1.ionoscloud.com).
Regions: Frankfurt, Berlin, Logrono (Spain), Lenexa (USA).
Addressing: virtual-hosted.
Notes: generate keys in the Data Center Designer under the Object Storage Key Manager^[10].

Gcore Object Storage

Gcore Object Storage is S3-compatible storage from Gcore, with a wide global edge footprint.

Endpoint format: regional S3 endpoint shown in the Gcore console (for example a s3.<region> style host).
Regions: many across Europe, the Americas, Asia, and the Middle East.
Addressing: virtual-hosted.
Notes: confirm the current endpoint host for your region in the Gcore console.

DreamHost DreamObjects

DreamObjects is DreamHost's S3-compatible object storage, built on Ceph.

Endpoint format: objects-<region>.dream.io (for example objects-us-east-1.dream.io).
Regions: US.
Addressing: virtual-hosted.
Notes: leave the region blank and set the endpoint, as with other Ceph-based services^[11].

RackCorp Object Storage

RackCorp Object Storage is an S3-compatible platform with anycast and many regional endpoints.

Endpoint format: <region>.s3.rackcorp.com (for example au-nsw.s3.rackcorp.com), or s3.rackcorp.com for the global anycast endpoint.
Regions: Australia (several), plus Asia, Europe, North America, and New Zealand.
Addressing: virtual-hosted.
Notes: useful where other providers lack a nearby region^[12].

cloudscale.ch

cloudscale.ch is a Swiss provider with S3-compatible object storage and strong data-residency guarantees.

Endpoint format: objects.<region>.cloudscale.ch (for example objects.rma.cloudscale.ch, objects.lpg.cloudscale.ch).
Regions: Rumlang and Lupfig, Switzerland.
Addressing: virtual-hosted.
Notes: all data stays in Switzerland^[13].

Fuga Cloud Object Store

Fuga Cloud is a European OpenStack-based provider with an S3-compatible object store.

Endpoint format: core.fuga.cloud.
Regions: the Netherlands.
Addressing: virtual-hosted.
Notes: confirm the endpoint and any newer hosts in the Fuga console.

Infomaniak Public Cloud

Infomaniak, a Swiss provider, offers S3-compatible object storage on its OpenStack-based Public Cloud.

Endpoint format: s3.<region>.infomaniak.cloud (for example s3.pub1.infomaniak.cloud).
Regions: Switzerland.
Addressing: virtual-hosted.
Notes: Swiss data residency and renewable-energy hosting.

Outscale OOS (Object Storage)

OUTSCALE, a Dassault Systemes brand, offers enterprise-grade S3-compatible object storage.

Endpoint format: oos.<region>.outscale.com (for example oos.eu-west-2.outscale.com).
Regions: Paris, New Jersey, California, SecNumCloud (Paris), Tokyo.
Addressing: virtual-hosted.
Notes: includes a SecNumCloud-qualified region for sensitive French workloads^[14].

Liara Object Storage

Liara is an Iranian platform with S3-compatible object storage.

Endpoint format: storage.iran.liara.space.
Regions: Iran.
Addressing: virtual-hosted.
Notes: leave the region blank and set the endpoint^[15].

ArvanCloud Object Storage

ArvanCloud offers S3-compatible object storage with regions in Iran.

Endpoint format: s3.<region>.arvanstorage.ir (for example s3.ir-thr-at1.arvanstorage.ir), also seen as s3.arvanstorage.com.
Regions: Tehran, Tabriz.
Addressing: virtual-hosted.
Notes: confirm the current endpoint host in the ArvanCloud console^[16].

BizflyCloud Simple Storage

Bizfly Cloud Simple Storage is an S3-compatible service with regions in Vietnam.

Endpoint format: <region>.ss.bfcplatform.vn (hn.ss.bfcplatform.vn for Hanoi, hcm.ss.bfcplatform.vn for Ho Chi Minh City).
Regions: Hanoi, Ho Chi Minh City.
Addressing: virtual-hosted.
Notes: match the region to the endpoint^[17].

Selectel and Servercore Object Storage

Selectel and its international brand Servercore offer S3-compatible object storage with triple-redundant storage.

Endpoint format: s3.<region>.storage.selcloud.ru (for example s3.ru-1.storage.selcloud.ru), plus s3.<region>.srvstorage.uz and s3.<region>.srvstorage.kz for Servercore in Central Asia.
Regions: St. Petersburg, Moscow, Kazakhstan, Uzbekistan.
Addressing: virtual-hosted (the provider recommends vHosted, not path-style).
Notes: confirm the endpoint for your region in the console^[18].

Fastly Object Storage

Fastly Object Storage is S3-compatible storage from the edge-cloud provider Fastly, with mandatory server-side encryption.

Endpoint format: <region>.object.fastlystorage.app (for example us-east.object.fastlystorage.app).
Regions: US East, US West, US Central, EU Central, EU South, EU West, UK East, JP Central, AU East.
Addressing: virtual-hosted.
Notes: server-side encryption is always on^[19].

Leaseweb Object Storage

Leaseweb offers S3-compatible object storage alongside its dedicated and cloud hosting.

Endpoint format: the regional S3 endpoint shown in the Leaseweb customer portal.
Regions: Europe, the US, and Asia-Pacific.
Addressing: virtual-hosted.
Notes: confirm the exact endpoint host in the Leaseweb portal.

Frequently Asked Questions

Why use a host's object storage instead of a specialist? Keeping storage with your compute reduces latency and often avoids cross-provider egress charges. If you already run servers at OVHcloud, Hetzner, or Vultr, their object storage is the path of least resistance.

Do these all use the AWS S3 API? Yes. Every provider here exposes an S3-compatible endpoint, so the same S3 tools and SDKs work by changing the endpoint and keys.

Can I move data from DigitalOcean Spaces to another host's object storage? Yes. Blober copies directly between S3 endpoints that support the normal object operations it uses. Set the source and destination and it transfers without staging a full copy on your disk.

Which of these connect to Blober without setup? DigitalOcean Spaces has a preconfigured connector. Every other provider on this page uses the generic S3-Compatible connector: paste the endpoint and keys.

Get Blober

Connect to any host's S3-compatible object storage by URL and move data between providers directly, without filling your local disk.

Download Blober at blober.io

S3-Compatible Storage Providers: The Complete List (90+ Providers)

Jun 28, 2026

This is a practical list of well-known S3-compatible storage providers. If a service speaks the Amazon S3 API, the same tools, SDKs, and apps that work with AWS S3 usually work with it by changing two things: the endpoint URL and the access keys. That portability is why so many S3-compatible services exist, from global clouds to storage you run on your own hardware.

We have grouped them into six categories so the list stays usable. Each provider below links to a detailed entry with its endpoint format, regions, addressing style, and notes. Whatever S3-compatible store you are searching for, there is a good chance it is on this page.

Every endpoint format here comes from the provider's own documentation, cross-checked against current S3 client references and the per-provider sources in the category pages^[1]. The tables show endpoint hostnames for readability. In Blober, enter the full URL with https:// unless you are connecting to a local/self-hosted server that explicitly uses http://.

What S3-Compatible Means Here

S3-compatible means the service exposes the same HTTP API as Amazon S3, so an S3 client points at it by URL. It does not mean the service is run by Amazon, and it does not promise every advanced feature is identical. For the full explanation, see What S3-Compatible Really Means.

For common S3 operations, the custom endpoint is the portability layer. Once your tool can set that endpoint, the provider becomes much easier to swap.

How Blober Connects to Any of These

Blober has preconfigured connectors for the most common S3 services (Amazon S3, Wasabi, Cloudflare R2, Backblaze B2, DigitalOcean Spaces, and Rabata). For everything else on this list, Blober has a single generic S3-Compatible connector: you paste the endpoint URL and keys, and Blober can use the common browse, upload, download, copy, and delete operations when the provider implements them.

That connector covers both addressing styles (bucket in the subdomain, or bucket in the path), single-bucket or full-account access, a custom region, and a free-text storage class, so it reaches the long tail of providers, self-hosted servers, and local test setups. You point Blober at a source and a destination and it copies between them directly, without staging a full copy on your disk.

The Six Categories

Category	What is in it	Open the list
Cloud object storage	S3 services from large platform clouds (Alibaba, Tencent, Huawei, IBM, Oracle, Google, Yandex, and more)	Cloud providers
Object storage specialists	Independent clouds built only for object storage (Wasabi, Backblaze, Storj, IDrive e2, Cubbit, and more)	Specialists
Hosting and VPS object storage	S3 storage bundled with a hosting or VPS platform (DigitalOcean, Linode, Vultr, OVHcloud, Hetzner, Scaleway, and more)	Hosting and VPS
Self-hosted and open source	S3 servers you run yourself (MinIO, Ceph, SeaweedFS, Garage, and more)	Self-hosted
Enterprise and on-premises	Appliances and enterprise platforms (NetApp StorageGRID, Dell ECS, Cloudian, Pure FlashBlade, and more)	Enterprise
Decentralized and Web3	Object storage backed by decentralized networks (Storj, Filebase, 4everland, and more)	Decentralized

Master Index of S3-Compatible Providers

Every provider, alphabetical, with its endpoint format. The endpoint is the host you give your S3 client; the bucket name is then added either as a subdomain or in the path. Replace any bracketed value (<region>, <account>, <namespace>) with your own. A check in the last column means Blober ships a preconfigured connector; everything else uses the generic S3-Compatible connector. When you enter one of these in Blober, include the URL scheme, for example https://s3.example.com.

Provider	Endpoint format	Preconfigured
Alibaba Cloud OSS	`oss-<region>.aliyuncs.com`
Amazon S3	`s3.<region>.amazonaws.com`	Yes
Apache Ozone	self-hosted S3 Gateway
ArvanCloud	`s3.<region>.arvanstorage.ir`
Backblaze B2	`s3.<region>.backblazeb2.com`	Yes
Baidu AI Cloud BOS	`s3.<region>.bcebos.com`
BizflyCloud	`<region>.ss.bfcplatform.vn`
Ceph (RADOS Gateway)	self-hosted
China Mobile Ecloud EOS	`eos-<city>-1.cmecloud.cn`
China Telecom CTYun OOS	`oos-<region>.ctyunapi.cn`
Cloudflare R2	`<account>.r2.cloudflarestorage.com`	Yes
Cloudian HyperStore	appliance or software
cloudscale.ch	`objects.<region>.cloudscale.ch`
Contabo Object Storage	`<region>.contabostorage.com`
Cubbit DS3	`s3.cubbit.eu`
DataCore Swarm	on-premises
Dell ECS / ObjectScale	on-premises
DigitalOcean Spaces	`<region>.digitaloceanspaces.com`	Yes
DreamHost DreamObjects	`objects-<region>.dream.io`
Exaba	self-hosted
Exoscale SOS	`sos-<zone>.exo.io`
Fastly Object Storage	`<region>.object.fastlystorage.app`
FileLu S5	`s5lu.com`
Filebase	`s3.filebase.io`
4everland	`endpoint.4everland.co`
Fuga Cloud Object Store	`core.fuga.cloud`
Garage	self-hosted
Gcore Object Storage	regional endpoint
Google Cloud Storage	`storage.googleapis.com`
Hitachi Content Platform	on-premises
Hetzner Object Storage	`<region>.your-objectstorage.com`
Huawei Cloud OBS	`obs.<region>.myhuaweicloud.com`
IBM Cloud Object Storage	`s3.<region>.cloud-object-storage.appdomain.cloud`
IDrive e2	`<region>.idrivee2-XX.com`
Impossible Cloud	`<region>.storage.impossibleapi.net`
Infomaniak Public Cloud	`s3.<region>.infomaniak.cloud`
Intercolo Object Storage	`de-fra.i3storage.com`
IONOS Cloud S3	`s3-<region>.ionoscloud.com`
JD Cloud Object Storage	`s3.<region>.jdcloud-oss.com`
Kingsoft Cloud KS3	`ks3-<region>.ksyuncloud.com`
Leaseweb Object Storage	regional endpoint
Leviia	`s3.leviia.com`
Liara	`storage.iran.liara.space`
Linode / Akamai	`<region>.linodeobjects.com`
MEGA S4	`s3.<region>.s4.mega.io`
MinIO	self-hosted
NAVER Cloud Object Storage	`<region>.object.ncloudstorage.com`
NetApp StorageGRID	on-premises
NetApp ONTAP S3	on-premises
Netease NOS	`nos-<region>.126.net`
Nutanix Objects	on-premises
OpenIO	self-hosted
OpenStack Swift (s3api)	self-hosted
Open Telekom Cloud OBS	`obs.<region>.otc.t-systems.com`
Oracle Cloud OCI	`<namespace>.compat.objectstorage.<region>.oci.customer-oci.com`
Outscale OOS	`oos.<region>.outscale.com`
OVHcloud Object Storage	`s3.<region>.io.cloud.ovh.net`
Petabox	`s3.<region>.petabox.io`
Pure Storage FlashBlade	on-premises
Qiniu Kodo	`s3-<region>.qiniucs.com`
Quantum ActiveScale	on-premises
Rabata	`s3.<region>.rabata.io`	Yes
RackCorp	`<region>.s3.rackcorp.com`
Riak CS	self-hosted
Scality RING / ARTESCA	on-premises
Scaleway Object Storage	`s3.<region>.scw.cloud`
Seagate Lyve Cloud	`s3.<region>.<account>.lyve.seagate.com`
SeaweedFS	self-hosted
Selectel	`s3.<region>.storage.selcloud.ru`
Servercore	`s3.<region>.storage.selcloud.ru`
SberCloud OBS	`obs.<region>.hc.sbercloud.ru`
Spectra Logic BlackPearl	on-premises
Storadera	`<region>.s3.storadera.com`
Storj	`gateway.storjshare.io`
Synology C2	`<region>.s3.synologyc2.net`
Tebi	`s3.tebi.io`
Telnyx Cloud Storage	`<region>.telnyxcloudstorage.com`
Tencent Cloud COS	`cos.<region>.myqcloud.com`
Tigris	`t3.storage.dev`
UpCloud Object Storage	`<region>.upcloudobjects.com`
US3 (UCloud)	`s3-<region>.ufileos.com`
VAST Data	on-premises
Versity Gateway	self-hosted
VK Cloud Object Storage	`hb.bizmrg.com`
Vultr Object Storage	`<region>.vultrobjects.com`
Wasabi	`s3.<region>.wasabisys.com`	Yes
Yandex Object Storage	`storage.yandexcloud.net`
Zadara	`<vsa-id>.zadarazios.com`
Zata	`idr01.zata.ai`

This index is a starting point. Providers add regions and change hostnames over time, so always confirm the exact endpoint in your provider's console before you connect. The detailed group articles explain each provider's regions, addressing style (path or subdomain), and any quirks.

Subdomain vs Path: The One Setting That Trips People Up

S3 supports two ways of putting the bucket name into the request:

Virtual-hosted (bucket in the subdomain): https://my-bucket.s3.example.com. This is the AWS default and what most hosted providers expect.
Path-style (bucket in the path): https://s3.example.com/my-bucket. This is what MinIO, Ceph, and most self-hosted servers expect.

If browsing or uploads fail against a self-hosted server, switch to path-style. In Blober, you pick the style implicitly by which endpoint field you fill in. Everything else is the same.

Frequently Asked Questions

How many S3-compatible storage providers are there? There is no official registry. This page indexes more than ninety well-known hosted providers, hosting platforms, self-hosted servers, enterprise systems, and decentralized gateways across six categories.

Is Azure Blob Storage S3-compatible? Not natively. Azure Blob uses its own API. The concepts map across (a container is like a bucket, a blob like an object), but an S3 tool needs a translation layer to talk to it. See How to Transfer AWS S3 to Azure Blob for moving data between the two.

Can I move data between two different S3-compatible providers? Yes. Because they share the core API, moving between them is usually a straightforward copy. Blober connects to a source and a destination and transfers directly between them. Check provider-specific feature gaps if your workflow depends on object lock, lifecycle, metadata edge cases, or server-side copy.

Do I need a preconfigured connector, or does the generic one work? The generic S3-Compatible connector works with S3-compatible endpoints that support the common operations Blober uses. The preconfigured connectors (S3, Wasabi, R2, B2, Spaces, Rabata) just save you from pasting the endpoint.

Which S3-compatible provider is the cheapest? It depends on your access pattern. Flat-rate, zero-egress providers win for data you read often; the lowest per-GB storage rate wins for cold archives. The pricing model (egress fees and minimum storage duration) usually matters more than the headline storage rate.

Get Blober

Connect to S3-compatible stores on this list and move data between them directly, without filling your local disk. Blober handles S3, Wasabi, R2, B2, Spaces, Rabata, and generic S3 endpoints by URL, plus non-S3 services like Azure Blob, Dropbox, and Google Drive.

Download Blober at blober.io

Self-Hosted and Open-Source S3-Compatible Storage (MinIO, Ceph, and More)

Jun 28, 2026

You do not need a cloud account to get an S3 endpoint. A whole category of open-source software gives you S3-compatible storage on your own hardware, in your own data center, or on a laptop for testing. These servers expose the same core S3 object-storage API shape as AWS, so the same tools can point at http://localhost:9000 as easily as at a cloud. This page lists the self-hosted and open-source S3 servers and how to connect each one.

This is one category in our complete list of S3-compatible storage providers. All of these connect to Blober through the generic S3-Compatible connector. Self-hosted servers almost always want path-style addressing (bucket in the path), so use the path endpoint field in Blober if browsing or uploads fail. The endpoint and compatibility notes below come from each project's own documentation, cross-checked against current S3 client references^[1]. The endpoint setup notes explain the difference between path-style and virtual-hosted addressing.

MinIO

MinIO is the most widely deployed self-hosted S3 server: a single Go binary that exposes an S3-compatible API, used for everything from local testing to large production clusters.

Endpoint format: http://<host>:9000 (for example http://localhost:9000).
Addressing: path-style.
Notes: for a quick local test, the default keys are often minioadmin / minioadmin. MinIO is also the reference target for "does this tool support custom S3 endpoints," and it is a common front end for other storage backends^[2].

Ceph (RADOS Gateway / RGW)

Ceph is an open-source distributed storage system; its RADOS Gateway (RGW) exposes an S3-compatible object interface on top of a Ceph cluster.

Endpoint format: https://<your-rgw-host> (your own gateway address).
Addressing: path-style is the safe default.
Notes: Ceph powers many commercial object stores (DreamObjects and Contabo among them). Leave the region blank and set the endpoint. Older Ceph (pre-Jewel) may need v2 signatures^[3].

SeaweedFS

SeaweedFS is a fast distributed storage system for blobs and files, with an S3-compatible gateway. It can also cache a remote S3 store with asynchronous write-back.

Endpoint format: http://<host>:8333 (for example http://localhost:8333).
Addressing: path-style.
Notes: lightweight and quick to stand up; good for edge and on-prem caching in front of a remote bucket^[4].

Garage

Garage, from Deuxfleurs, is a lightweight, self-hosted S3-compatible object store designed for geo-distributed deployments on modest hardware.

Endpoint format: http://<your-garage-host>:3900 (your own configured address).
Addressing: path-style.
Notes: built for resilience across home-server-grade nodes; a popular choice for self-hosters and small co-ops^[5].

Zenko CloudServer

Zenko CloudServer (formerly Scality S3 Server) is an open-source Node.js S3 server that can store locally or proxy to cloud backends.

Endpoint format: http://<host>:8000 (your own configured address).
Addressing: path-style.
Notes: part of the wider Zenko project, which targets multi-cloud data management with an S3 front end^[6].

Versity Gateway (VersityGW)

Versity Gateway is an open-source S3 gateway with pluggable backends, letting you put an S3 API in front of a POSIX file system or other stores.

Endpoint format: http://<your-gateway-host>:7070 (your own configured address).
Addressing: path-style.
Notes: useful for exposing existing file storage (including tape-backed archives) over S3^[7].

s3proxy

s3proxy is an open-source proxy that presents an S3 API backed by a local file system or by other cloud stores, including Azure Blob and Google Cloud Storage.

Endpoint format: http://<your-proxy-host>:80 (your own configured address).
Addressing: path-style.
Notes: a common way to give non-S3 backends an S3 face. If you specifically need S3 in front of Azure, this is one of the standard tools^[8].

Apache Ozone

Apache Ozone is a distributed object store for big-data and Kubernetes workloads, with an S3 Gateway component.

Endpoint format: the S3 Gateway address you deploy (commonly port 9878).
Addressing: path-style.
Notes: designed to scale to billions of objects alongside Hadoop and Spark^[9].

OpenStack Swift (with the s3api middleware)

OpenStack Swift is the object storage component of OpenStack. With the s3api middleware enabled, it accepts S3 API calls in addition to its native Swift API.

Endpoint format: your Swift proxy address with s3api enabled (your own deployment).
Addressing: path-style.
Notes: many regional and academic clouds run Swift; if yours has s3api turned on, you can connect over S3^[10].

OpenIO

OpenIO is an open-source object storage platform (now part of OVHcloud) with an S3-compatible gateway.

Endpoint format: your OpenIO S3 gateway address (your own deployment).
Addressing: path-style.
Notes: designed for grid-style scaling on heterogeneous hardware^[11].

LeoFS

LeoFS is an open-source distributed object store with an S3-compatible interface.

Endpoint format: your LeoFS gateway address (your own deployment).
Addressing: path-style.
Notes: confirm the gateway endpoint and credentials from your LeoFS configuration^[12].

Riak CS

Riak CS (Cloud Storage) layers S3-compatible object storage on top of a Riak key-value cluster.

Endpoint format: your Riak CS proxy address (your own deployment).
Addressing: path-style.
Notes: suited to multi-tenant, highly available deployments^[13].

Rclone Serve S3

Rclone can serve any of its remotes over the S3 protocol with rclone serve s3, turning anything rclone supports into a temporary S3 endpoint.

Endpoint format: http://<host>:8080 (your own configured address).
Addressing: path-style.
Notes: handy for bridging an odd backend to S3 tools for a one-off transfer^[14].

Exaba

Exaba is an on-premises, S3-compatible storage server for service providers and enterprises, with a free container edition.

Endpoint format: http://<host>:9000 (your own configured address; the admin runs on a separate port).
Addressing: path-style.
Notes: set up the container through the admin interface first, then use the S3 service port^[15].

A Note on Gateways: S3 in Front of Something Else

Several tools here (s3proxy, Versity Gateway, Zenko, MinIO, and SeaweedFS) can act as gateways: they present an S3 API in front of a different backend, such as a local file system, tape, or even a non-S3 cloud like Azure Blob. If your goal is to make a non-S3 store look like S3, a gateway is the usual answer. If your goal is simply to move data between an S3 store and a non-S3 store, Blober already bridges S3, Azure Blob, Dropbox, Google Drive, and more directly, with no gateway to run.

Frequently Asked Questions

Which self-hosted S3 server should I start with? MinIO, for most people. It is a single binary, well documented, and the de facto standard for a local or on-prem S3 endpoint. Ceph RGW suits larger clusters; Garage and SeaweedFS suit lightweight or distributed setups.

Why do uploads fail against my self-hosted server? Almost always because the tool is using virtual-hosted (subdomain) addressing. Self-hosted servers want path-style. In Blober, use the path endpoint field, for example http://localhost:9000.

Can Blober connect to a server on my own machine? Yes. Point it at http://localhost:9000 (or your server's address) with path-style addressing and your keys, and browse to confirm.

Can I put an S3 API in front of Azure or Google Drive? You can with a gateway like s3proxy. But to just move data between those services and an S3 store, Blober connects to each one natively, so you do not need a gateway.

Get Blober

Connect Blober to MinIO, Ceph, SeaweedFS, Garage, or any self-hosted S3 server by URL, and move data between your own storage and the cloud directly, without filling your local disk.

Download Blober at blober.io

Large Cloud Transfers: When a Desktop App Is the Right Tool

Jun 27, 2026

Moving a few files is easy anywhere. Moving terabytes is a question of where the work runs, and the answer depends on the job. For most transfers, a desktop app on your own machine beats a browser-based or hosted tool. For a single bulk move of a very large dataset into one provider, that provider's own migration service can be faster. This page sorts out which is which.

The Ceiling Inside a Browser Tab

A browser-based transfer happens inside a tab, and a tab is a small, supervised space. It has a memory budget, it can be paused or discarded when you switch away, and it usually holds only a handful of connections. For a dozen photos this is invisible. For a media library it is the wall the transfer hits: the tab slows, the laptop heats up, and a job that should run for an hour cannot stay awake long enough to finish.

The Timeout and the Bandwidth Bill

Hosted services move work onto their own machines, which sounds like an upgrade until you look at the path. The data has to travel from the source into their infrastructure, and then back out to the destination. You wait for both legs, and the company pays for the bandwidth in the middle. That cost does not disappear. It comes back as a subscription, a per-GB charge, or a monthly transfer cap.

Server functions also run on a clock. AWS Lambda, for example, documents a maximum function timeout of 900 seconds, or 15 minutes^[1]. Many serverless jobs are designed to finish in seconds or minutes, not the hours a large migration needs. When the clock runs out mid-file, the job has to be broken into pieces and stitched back together; if that orchestration is wrong, gaps and duplicates are the failure mode.

What a Desktop Process Can Actually Do

A desktop app is not boxed in the same way. It can use the memory and processor your machine actually has, open many transfers in parallel, and keep running for as long as the work takes. If the connection drops, Blober can retry failed records; if you stop and rerun a workflow, skip-existing keeps completed files from being sent again.

Blober also streams rather than stages. Instead of saving a full copy of every file to your disk and uploading it again, it flows the data through and writes it straight to the destination, so disk space stops being the limit. The difference between a streamed transfer and download-and-reupload is small for ten files and decisive for ten thousand, which is covered in Server-to-Server vs Download-and-Reupload.

Your Connection, Your Rules

Because the transfer runs from your machine, it uses your own connection and your own IP address. There is no shared relay address for a provider to throttle or block for every user of the tool at once. A hosted service, by contrast, reaches every provider from the same pool of company servers, which is the traffic pattern most likely to attract shared-service limits. Your own account and API limits still apply either way.

When a Provider's Own Migration Wins

For a single move of a very large dataset, tens of terabytes or tens of millions of files, into one destination, the fastest path is often not your machine at all. Some providers run datacenter-to-datacenter migration, where the data never makes the round trip through your home or office connection. Backblaze, for example, lists assisted data migration as available for B2 Cloud Storage^[2], and what can take weeks streaming through your own link can take far less over a datacenter one. If you are doing one big move into one provider and you can grant that service access, start there. There is a worked 25 TB example in Migrating 100M+ Files from DigitalOcean Spaces to Backblaze B2.

This is not a flaw in any one tool. Any client-side transfer, a desktop app or a command-line tool alike, has to pull the data down to your machine and push it back up, which is two trips through your connection. For the very largest one-time moves, skipping those trips is the whole point.

Where a Desktop App Is the Right Tool

A provider migration helps in a narrow case: one big move, into one provider that offers it, when you can hand that provider access. Outside that case, a desktop app like Blober is usually the better fit, and for some teams it is the only acceptable one.

When you cannot share credentials with anyone. Assisted migration means granting a third party access to your storage. For regulated data, client data under NDA, or anything inside a SOC 2 or ISO 27001 boundary, that is often ruled out. A team in that position cannot hand its keys to a destination provider's migration service any more than it could hand them to a relay. Blober keeps your credentials on your own machine and adds no third party to the path, so the transfer stays inside the controls you already have. See How Blober Works.
When the move is ongoing, not one-time. Migration services are built for a single bulk lift. New uploads, folder moves, periodic refreshes, and re-running a saved job are what a desktop tool is for.
When no assisted path exists. Provider migration almost always moves data into that one provider. Moving between two arbitrary providers, or pulling a library out of a consumer cloud like GoPro Cloud or Google Photos, has no assisted option. A desktop app that speaks each provider directly is the practical route.
When the scale is moderate. For gigabytes or a few terabytes, arranging an assisted migration is more trouble than it is worth. A desktop app just runs.
When you want to watch it happen. A desktop transfer stays visible and under your control, so you can pause, resume, and verify on your own machine.

What a Desktop App Does and Does Not Solve

A desktop app is not magic, and it is not always the fastest option. The bytes travel over your own connection, so a slow link is still slow, and for the very largest one-time moves a provider's datacenter migration can beat it outright. What a desktop app gives you is control and privacy: no relay, no third party holding your keys, and the reach to move between any providers you connect, including the ones with no migration service at all.

Frequently Asked Questions

Is a desktop app the fastest way to move a very large dataset? Not always. For a one-time move of tens of terabytes into a single provider that offers datacenter migration, that route is usually faster, because it skips the trip through your connection. A desktop app wins on control, privacy, ongoing use, and moves between providers that have no assisted option.

What if I cannot give a third party access to my storage? Then a desktop app is the right choice. Blober uses your credentials on your own machine and adds no outside party to the transfer, so it fits inside compliance boundaries that rule out both hosted relays and assisted migration services.

Does the data still go through my computer? Yes. A desktop tool streams the data through your machine to the destination. The point is that it does not also route through a separate company's servers, and it does not save a second full copy to your disk.

What happens if my transfer is interrupted? Blober retries failed records during a run. If you stop and rerun the workflow, skip-existing avoids sending files already written at the destination, so the job continues at file level instead of starting the whole migration again.

Next Steps

For a very large one-time move, read Migrating 100M+ Files from DigitalOcean Spaces to Backblaze B2.
See how the pieces fit in How Blober Works.
Read the mechanic behind direct transfers in Server-to-Server vs Download-and-Reupload.
Ready to move your files? Get Blober.

Articles

Alibaba Cloud OSS (Object Storage Service)

Tencent Cloud COS (Cloud Object Storage)

Huawei Cloud OBS (Object Storage Service)

IBM Cloud Object Storage (COS)

Oracle Cloud Infrastructure (OCI) Object Storage

Google Cloud Storage (XML / S3 Interoperability)

Yandex Object Storage

VK Cloud Object Storage (formerly Mail.ru Cloud Solutions)

Baidu AI Cloud BOS (Baidu Object Storage)

JD Cloud Object Storage (OSS)

China Mobile Ecloud EOS (Elastic Object Storage)

Kingsoft Cloud KS3

China Telecom CTYun OOS

NAVER Cloud Object Storage

Open Telekom Cloud OBS

SberCloud OBS

Netease NOS (Netease Object Storage)

A Note on Microsoft Azure Blob Storage

Frequently Asked Questions

Related Guides

Get Blober

DigitalOcean Spaces

Linode / Akamai Object Storage

Vultr Object Storage

OVHcloud Object Storage

Hetzner Object Storage

Scaleway Object Storage

Contabo Object Storage

Exoscale SOS (Simple Object Storage)

UpCloud Object Storage

IONOS Cloud S3 Object Storage

Gcore Object Storage

DreamHost DreamObjects

RackCorp Object Storage

cloudscale.ch

Fuga Cloud Object Store

Infomaniak Public Cloud

Outscale OOS (Object Storage)

Liara Object Storage

ArvanCloud Object Storage

BizflyCloud Simple Storage

Selectel and Servercore Object Storage

Fastly Object Storage

Leaseweb Object Storage

Frequently Asked Questions

Related Guides

Get Blober

What S3-Compatible Means Here

How Blober Connects to Any of These

The Six Categories

Master Index of S3-Compatible Providers

Subdomain vs Path: The One Setting That Trips People Up

Frequently Asked Questions

Related Guides

Get Blober

MinIO

Ceph (RADOS Gateway / RGW)

SeaweedFS

Garage

Zenko CloudServer

Versity Gateway (VersityGW)

s3proxy

Apache Ozone

OpenStack Swift (with the s3api middleware)

OpenIO

LeoFS

Riak CS

Rclone Serve S3

Exaba

A Note on Gateways: S3 in Front of Something Else

Frequently Asked Questions

Related Guides

Get Blober

The Ceiling Inside a Browser Tab

The Timeout and the Bandwidth Bill

What a Desktop Process Can Actually Do

Your Connection, Your Rules

When a Provider's Own Migration Wins

Where a Desktop App Is the Right Tool