STANDARD_IA) for infrequently accessed data, at a lower per-GiB rate than active storage. Archive is in limited availability. To request access for your organization, contact CoreWeave support.
Overview
Archive suits long-term retention of training datasets, model checkpoints, processed datasets, experiment artifacts, and logs. Unlike most archival storage products, Archive preserves the workflows you already use:- Objects keep the same bucket and object address.
- Applications continue to use the S3-compatible API.
- Reads don’t require a restore or rehydration step.
- Reads don’t incur separate retrieval or early-deletion fees under the default configuration.
STANDARD remains the default for new objects. Hot, Warm, and Cold are usage-based billing classifications within STANDARD, not separate storage classes.
Move data into Archive
Three paths move data into Archive:
Automated Archive is a control-plane setting, not an S3 Lifecycle transition. AI Object Storage doesn’t support
Transition lifecycle actions.
Things to know
- Reading an archived object doesn’t return it to
STANDARD. Move objects out of Archive explicitly before running a latency-sensitive or high-throughput workload against them. - Archive performance is provisioned per geo. Coordinate bulk reads and large-scale migrations with CoreWeave support in advance.
- Archive data can be stored outside the bucket’s data center. If you have data residency requirements, review the placement model before you enable Archive.
- Configuring Automated Archive requires the
cwobject:ConfigureBucketArchiveandcwobject:GetBucketInfoactions.
Learn more
- For the full guide, including performance expectations, Automated Archive configuration, and billing details, see Use the Archive storage class.
- For
STANDARD_IAcapacity quotas and storage capacity alerts, see Manage quotas and limits. - To configure Automated Archive with Terraform, see
coreweave_object_storage_bucket_settings.