Download Objects

Learn about downloading objects

You can download Supabase Storage files can in a variety of ways depending on your use case.

From the Dashboard

You can download individual files directly from the Storage > Files section of the Dashboard by browsing to your bucket and selecting files.

Using the Supabase CLI

You can use the supabase storage commands to list and copy objects from your linked project by passing --remote.

See the CLI reference for available commands.

Using an S3-compatible client

Supabase Storage exposes an S3-compatible endpoint, which means you can use tools like Cyberduck (via a connection profile), AWS CLI, or rclone to browse and download your files.

To connect:

Go to Storage > Configuration > S3 section of the Dashboard.
Enable the S3 protocol and generate an access key + secret key pair (Save it immediately as it is shown once).
Use your S3-compatible tool with the endpoint and credentials from the Dashboard.

See the S3 authentication doc for full connection details.

For downloading a large number of objects, using an S3-compatible tool like rclone or Cyberduck is significantly more efficient than downloading files individually.

Using migration scripts

For bulk downloads or project migrations, you can use the scripts from our migration docs:

Node.js script — see "Migrating storage objects" in the migration guide
Google Colab script — see "Migrate storage objects to new project's S3 storage" in the dashboard restore guide

File metadata

File metadata is stored separately from the actual files. It lives in the storage.buckets and storage.objects tables in your Postgres database. If you need a complete backup (files + metadata), see the backup and restore guide.