SupportConsoleDevelopersStart a trialContact

Chapter 5. Create and use custom endpoints in the playground

Important

Custom endpoints is a Technology Preview feature only. Technology Preview features are not supported with Red Hat production service level agreements (SLAs) and might not be functionally complete. Red Hat does not recommend using them in production. These features provide early access to upcoming product features, enabling customers to test functionality and provide feedback during the development process.

For more information about the support scope of Red Hat Technology Preview features, see Technology Preview Features Support Scope.

You can create custom endpoints to use models that are not deployed in your project namespace. Custom endpoints let you access models deployed in another namespace on the same cluster, or models hosted by external third-party providers such as OpenAI, Anthropic, or AWS.

After you create a custom endpoint, the model appears on the AI asset endpoints page and is available for selection in the playground.

Warning

When you connect to an external third-party provider, data from the responses API, including RAG context, MCP tool results, and user input, is sent outside your cluster. Ensure that your organization’s security policies allow this before creating an external endpoint.

Prerequisites

  • You have configured a playground for your project.
  • Your platform engineer has enabled the custom endpoints feature for your project. For more information, see Enable custom endpoints for the playground.
  • You can create configmaps and secrets in your namespace.
  • If you are connecting to a model in another namespace, you have the internal API endpoint URL and an authorized access token.
  • If you are connecting to an external provider, you have the provider’s API endpoint URL and a valid API key.

Procedure

  1. From the OpenShift AI dashboard, click Gen AI studio AI asset endpoints.
  2. Click Create endpoint.

  3. In the form, enter the following details:

    Model type
    Specifies whether the model is an inferencing model or an embedding model. Inferencing models generate text responses and are used in the playground. Embedding models convert text to vectors and are used in RAG pipelines.
    Model ID
    Specifies the exact model identifier from your provider. This must match the provider’s model ID exactly.
    Display name
    Specifies a user-friendly name for the model, shown in tables and selectors instead of the model ID.
    Embedding dimension
    For embedding models only. Specifies the output vector size for the embedding model.
    URL
    Specifies the API endpoint for the model service. For a model in another namespace, use the internal cluster URL, for example https://<service-name>.<namespace>.svc.cluster.local. For an external provider, use the provider’s API URL.
    Token
    Specifies the API key or access token required to authenticate with the endpoint. The credential is stored as a Kubernetes Secret and is shared at the project level.
    Use case
    Optional. Specifies what the model is best suited for, to help other users in the project identify the model’s purpose.
  4. Optional: Click Verify model to send a test request to the endpoint. The verification checks that the endpoint is reachable and returns an OpenAI-compatible response.

  5. Click Create.

    The custom endpoint appears on the AI asset endpoints page.

  6. Click Add to Playground to make the model available in the playground.
  7. To use the model, click Try in Playground, or select the model from the models dropdown in the playground.

Verification

  • The custom endpoint appears on the AI asset endpoints page with the correct endpoint type.
  • You can select the model in the playground and receive inference responses.
Red Hat logoGithubredditYoutubeTwitter

Learn

Try, buy, & sell

Communities

About Red Hat

We deliver hardened solutions that make it easier for enterprises to work across platforms and environments, from the core datacenter to the network edge.

Making open source more inclusive

Red Hat is committed to replacing problematic language in our code, documentation, and web properties. For more details, see the Red Hat Blog.

About Red Hat Documentation

We help Red Hat users innovate and achieve their goals with our products and services with content they can trust. Explore our recent updates.

Legal Notice

Theme

© 2026 Red HatBack to top