SupportConsoleDevelopersStart a trialContact

Chapter 1. Playground overview

The generative AI (gen AI) playground is an interactive environment within the Red Hat OpenShift AI dashboard where you can prototype and evaluate foundation models, custom models, and Model Context Protocol (MCP) servers before you use them in an application.

Important

Gen AI playground 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 test different configurations, including retrieval augmented generation (RAG), to determine the right assets for your use case. After you find an effective configuration, you can retrieve a Python template that serves as a starting point for building and iterating in a local development environment.

The playground interface consists of a chat area and a Configure panel. The Configure panel is organized into tabs:

  • Model: Select a model and adjust parameters such as temperature and streaming. Optionally, select a transcription model for audio upload.
  • Prompt: Write, save, and load system instructions.
  • Knowledge: Upload files for RAG or select an external vector store.
  • MCP: Connect to Model Context Protocol servers and authorize access to their tools.
  • Guardrails: If your cluster administrator has enabled guardrails, configure input and output guardrail settings.
Note

Chat history is not preserved if you refresh your browser or end your session. However, you can save your playground configuration, including model selection, inference parameters, prompt, knowledge sources, and MCP servers, as a named reusable agent. Prompts that you save through the Prompt tab are stored in MLflow and also persist across sessions. Saving playground agents is a Developer Preview feature. For more information, see Saved agents in the gen AI playground.

The playground supports the following workflows:

  • Chat with foundation models and custom-deployed models, including models from external endpoints.
  • Test prompt engineering with RAG by uploading documents or selecting external vector stores as knowledge sources.
  • Compare model responses side by side by opening multiple chat panes.
  • Save and version system instructions as reusable prompts.
  • Upload images to vision-enabled models, upload audio files for speech-to-text transcription, and attach documents for RAG context through the attach menu. For more information, see How multimodal input extends model evaluation.
  • Authorize and interact with MCP servers and their tools.
  • (Technology Preview) Enable distributed tracing to view inline performance metrics and inspect traces for each chat response.
  • (Developer Preview) Save and load playground configurations as named reusable agents.
  • View your playground configuration as Python code that you can copy for use in a local development environment.
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