Skip to content
Storage

Storage features

The S3 you know, minus the bill you don't

Bucket and object operations, multipart uploads, presigned URLs and scoped keys — all over the S3 API, all on NVMe, all without egress or request fees.
Create a bucketRead the docs

Free tier: 2 buckets and 5 GB. No card required.

Compatibility

Works with the tools you already use

Set the endpoint to storage.edge.network, turn on path-style addressing and use Edge keys. That's the whole migration for most code.
  • AWS CLI
  • AWS SDK (JS, Python, Go…)
  • boto3
  • rclone
  • s3cmd
  • Cyberduck
  • MinIO Client

Bucket operations

  • CreateBucket
  • DeleteBucket
  • ListBuckets
  • HeadBucket

Object operations

  • PutObject
  • GetObject
  • HeadObject
  • DeleteObject
  • ListObjectsV2
  • CopyObject

Multipart uploads

  • CreateMultipartUpload
  • UploadPart
  • CompleteMultipartUpload
  • AbortMultipartUpload
  • ListParts
  • ListMultipartUploads

Signing

  • AWS Signature V4
  • Presigned GET
  • Presigned PUT

Access control

Give every app only the keys it needs

Create as many access keys as you like. Make a key read-only or read-write, and scope it to specific buckets — anything outside the scope returns AccessDenied.
  • Read-only keys for analytics and public-facing reads
  • Bucket-scoped keys for each app and environment
  • Revoke instantly; secrets shown once at creation
Access key docs

Lifecycle rules

Temporary files that clean up after themselves

Add expiry rules per bucket, optionally filtered by prefix. Rules run hourly and delete objects older than the threshold — so tmp uploads, logs and build artefacts don't pile up.
  • Several rules per bucket, each with its own prefix
  • Set in the console or with one API call
  • Pairs well with $0 requests: expiring costs nothing
Bucket docs

Multipart uploads

Big files, uploaded in parallel and resumed on failure

Split large files into parts, upload several at once and retry only the part that fails. The AWS CLI and SDK helpers do it for you automatically.
  • Objects up to 5 TB
  • Parts from 5 MB to 5 GB, up to 10,000 per upload
  • Per-part ETags for integrity
Multipart docs

Metrics

Know what's stored, and what's popular

Every bucket reports storage, request volume and bandwidth hourly, plus a breakdown by file type and a live list of the most-requested objects.
  • 24h, 7d and 30d views, hourly resolution
  • Hourly data kept for 90 days
  • Available in the console and via API
Metrics docs

And more

Also on every bucket

  • CDN origin

    Use a private bucket as an Edge CDN origin and cache objects at 988 locations.
    Edge CDN

  • Public buckets

    One toggle for unauthenticated GET and HEAD, with CORS headers for browsers.

  • Console uploads

    Drag and drop files or whole folders, with progress, retries and in-browser previews.

  • Per-account isolation

    Each account has dedicated storage and metadata databases, for security and performance.

  • ZIP downloads

    Select up to 100 objects in the console or API and download them as one archive.

  • Activity log

    Every bucket keeps an event log of uploads, deletes and settings changes.

Specifications

Limits and connection details

Connection
endpoint        https://storage.edge.network
region          us-east-1   # any valid string
forcePathStyle  true
ResourceLimit
Single PUT (S3 API)5 GB
Object size (multipart)5 TB
Multipart part size5 MB – 5 GB
Parts per upload10,000
Console upload per request1 GB
Objects per bucketUnlimited
Presigned URL lifetimeUp to 7 days
Buckets on the free tier2

Point your SDK at Edge

Create a bucket and a key, change one endpoint, and your existing code is serving with zero egress.

Create a bucketSee pricing

Free to sign up. Nothing on Edge silently starts billing.