Get Started with AI Object Storage
Here's a step-by-step guide to set up your environment and start using CoreWeave AI Object Storage for your projects. To get started with CoreWeave AI Object Storage, you need:
- Administrative access to your CoreWeave organization
- A Cloud Console token with
admin
permissions
This guide outlines the steps needed to create Object Storage Access Keys and buckets; set policies for access, lifecycle, and auditing; and use the objects and buckets to store and retrieve data.
1. Create Access Keys
CoreWeave AI Object Storage uses Access Keys to authenticate users. These keys can be persistent if generated via the Cloud Console, or temporary if generated using SAML assertions.
At a minimum, S3:CreateBucket
is required to create buckets in CoreWeave AI Object Storage.
See Concepts: Access Control and Reference: Bucket Policy to learn more.
Use SAML/SSO assertions
To create time-limited Access Keys for CoreWeave AI Object Storage with SAML/SSO assertions:
- Enable SAML/SSO for your organization. You'll need information about your specific IdP, such as the IdP's SSO URL, unique Entity ID, and security certificate.
- Create a time-limited Access Key with a SAML assertion by following the steps in How-To: Create Key with a SAML assertion.
After creating the SAML/SSO Access Key, create a bucket to store data.
Use Cloud Console tokens
To create either permanent or time-limited Access Keys for CoreWeave AI Object Storage with Cloud Console tokens, follow the steps in How-To: Create Key with a Cloud Console token. The Cloud Console token used to create the Access Key must have write
permissions for Object Storage to allow users to upload data. See Create an API Access Token for details.
2. Create a bucket
Buckets are the primary organizational unit for CoreWeave AI Object Storage used to store object data. Buckets can be created with the Object Storage API. See How-To: Manage Buckets to learn how to create buckets with popular S3-compatible tools using either the primary object storage endpoint or the accelerated LOTA endpoint.
3. Set policies
Before uploading data, you may want to set Bucket Policies for your organization and buckets. Policies are used to control access to data, control lifecycles, and configure auditing.
See these sections for more information:
- Concepts: Bucket Policies
- How-To: Manage Bucket Policies
- Reference: Bucket Policy
- Reference:
EnsureAccessPolicy
action
4. Use objects and buckets
CoreWeave AI Object Storage is compatible with the S3 API, so you can use any S3-compatible tool to manage your data. See How-To: Manage Objects to learn how to use CoreWeave AI Object Storage with s3cmd, Boto3, or the AWS CLI.
CoreWeave AI Object Storage is designed to work with LOTA (Local Object Transfer Accelerator), a first-of-its-kind intelligent proxy that provides a highly efficient, local gateway to CoreWeave AI Object Storage. See Concepts: LOTA for details about how LOTA works.
CoreWeave AI Object Storage has default quota limits. See Manage quota limits for more information.
API reference and compatibility information
- For API reference for AI Object Storage, refer to the API documentation.
- For S3 API compatibility reference, refer to the S3 compatibility documentation.