Oracle Cloud Infrastructure Documentation

Updating a Dedicated AI Cluster

Update the name or description of a dedicated AI cluster in OCI Generative AI. If the cluster is a hosting cluster, you can also add units (model replicas) to it.

When you edit a dedicated AI cluster, you can also update its tags. For instructions, see Updating a Tag for a Single Resource. For more information about tagging, see Resource Tags.

For a hosting cluster, adding model replicas increases inference throughput and requires additional dedicated AI units. Adding replicas doesn't create model endpoints. To create another access point for a compatible model on the cluster, create a model endpoint.

    1. On the Dedicated AI clusters list page, select the dedicated AI cluster that you want to work with. If you need help finding the list page for dedicated AI clusters, see Listing Dedicated AI Clusters.
    2. On the details page, select Edit.
    3. Update the name and the description.
    4. For hosting clusters, add or remove units if needed.
      Add model replicas to increase inference throughput. Creating additional endpoints alone doesn't increase throughput.
    5. Select Update.

  • Use the dedicated-ai-cluster update command and required parameters to update a dedicated cluster:

    oci generative-ai dedicated-ai-cluster update 
--dedicated-ai-cluster-id <dedicated-ai-cluster-OCID> [OPTIONS]

    For a complete list of parameters and values for CLI commands, see the CLI Command Reference.

  • Run the UpdateDedicatedAiCluster operation to update a dedicated cluster.