Skip to main content
CoreWeave AI Object Storage has the following quotas and limits by default.

​
Quotas

Object Storage tracks capacity quotas separately for each storage class. The quota for STANDARD applies to the Hot, Warm, and Cold tiers combined. The quota for STANDARD_IA applies to the Archive tier, if your organization uses this tier.
QuotaDefault value
Default STANDARD capacity (per region, per account)100 TiB
Default STANDARD_IA capacity (per region, per account)0 TiB
Default number of buckets per account1,000
Default LOTA cache capacity per Node1 TiB
When usage in a storage class reaches that class’s capacity limit, write operations to that storage class are blocked. To request a quota increase, contact CoreWeave support and include:
  • Your organization name and Org ID, found on the Settings page.
  • The region and zone.
  • The storage class (STANDARD or STANDARD_IA) the request applies to.
  • The requested object storage quota (TiB).
  • A brief description of the workloads and expected storage growth (Optional). For example, new clusters or changes over the next 30 to 90 days.

​
Limits

The following limits are technical constraints on individual objects and multipart uploads.
LimitDefault value
Maximum part size5 GiB
Minimum part size15 MiB
Maximum object size48.828125 TiB
Maximum parts per multipart upload10,000
Lifecycle rules per configuration1,000
1 The last object part of a multipart upload can be less than 5 MiB in size.

​
Storage capacity alerts

Object Storage sends alerts when your usage approaches or is projected to exceed your quota, giving you time to act before write operations are blocked. Alerts fire independently for each storage class quota. If your organization has both STANDARD and STANDARD_IA quotas, you receive separate alerts for each. The following table lists each alert and the condition that triggers it:
Alert nameConditionSeverity
CAIOSQuotaProjectedNearLimitUsage is projected to exceed 90% of quota within 7 daysWarning
CAIOSQuotaNearLimitCurrent usage exceeds 90% of quotaWarning
CAIOSQuotaProjectedLimitUsage is projected to exceed 100% of quota within 7 daysCritical
When multiple alerts fire at the same time for the same storage class and zone, only the highest-severity alert fires and the others are suppressed. Each notification covers a single storage class in a single zone, so multiple zones or storage classes approaching the limit produce separate notifications. Alerts are delivered through your configured integration. To set up Slack or webhook delivery, see CoreWeave Alerts.

​
PromQL for custom monitoring

You can use the following PromQL expressions against your Grafana datasource to build your own dashboards or alerting rules. The examples query the STANDARD storage class. To query the STANDARD_IA storage class instead, replace storage_class="STANDARD" with storage_class="STANDARD_IA" in each block.

​
Current usage as a percentage of quota

label_replace(
  sum(cwobject_bucket_usage{
    measurement_type="MEASUREMENT_TYPE_USAGE_BYTES",
    storage_class="STANDARD"
  }) by (cluster_org, bucket_zone, storage_class),
  "zone", "$1", "bucket_zone", "(.*)"
)
/ on(cluster_org, zone, storage_class)
label_replace(
  cwobject_quota_info{
    active="true",
    measurement_type="MEASUREMENT_TYPE_USAGE_BYTES",
    storage_class="STANDARD"
  },
  "zone", "$1", "quota_zone", "(.*)"
)
* 100

​
Projected usage in 7 days as a percentage of quota

The following example uses predict_linear over a 1-week lookback: 
predict_linear(
  (
    label_replace(
      sum(cwobject_bucket_usage{
        measurement_type="MEASUREMENT_TYPE_USAGE_BYTES",
        storage_class="STANDARD"
      }) by (cluster_org, bucket_zone, storage_class),
      "zone", "$1", "bucket_zone", "(.*)"
    )
  )[1w:],
  7 * 24 * 3600
)
/ on(cluster_org, zone, storage_class)
label_replace(
  cwobject_quota_info{
    active="true",
    measurement_type="MEASUREMENT_TYPE_USAGE_BYTES",
    storage_class="STANDARD"
  },
  "zone", "$1", "quota_zone", "(.*)"
)
* 100
Last modified on May 29, 2026