Red Hat OpenShift Lightspeed

Red Hat OpenShift Lightspeed brings AI assistance directly into the Red Hat OpenShift web console, helping technical teams answer questions, troubleshoot workloads, and visualize metrics. You can use natural language prompts to get step-by-step guidance, receive AI-generated diagnostic summaries, and access on-demand cluster insights without leaving your workspace. Answers are drawn from official Red Hat documentation and powered by your organization’s selected large language model (LLM) provider.

Red Hat OpenShift Lightspeed is designed to answer questions about Red Hat OpenShift and additional components, such as OpenShift Virtualization, OpenShift GitOps, OpenShift Pipelines, and Red Hat Advanced Cluster Management for Kubernetes. Answers to knowledge questions include links to the product documentation so you can see where the information came from and learn more.

OpenShift Lightspeed can also help with troubleshooting. You can attach an alert message or your YAML file to the chat window and ask what you should do about it. If you have an issue with a pod, you can submit the status information of the pod you’re viewing and ask for help solving the problem.

User-initiated troubleshooting mode

When an alert fires, site reliability engineers (SREs) can use OpenShift Lightspeed directly from the console to investigate rather than checking logs manually. OpenShift Lightspeed’s troubleshooting mode gathers live cluster data—such as Prometheus metrics, memory use, restart counts, and pod logs—to summarize symptoms, evidence, and root causes alongside specific fix commands. Human operators remain in control of which commands to run.

OpenShift Lightspeed’s agentic capabilities go beyond passive diagnosis. They take action. When an alert fires or a cluster upgrade is planned, OpenShift Lightspeed initiates an automated agentic run—a multi-phase workflow that analyzes symptoms, traces root causes, catches conflicting issues, and delivers precise, actionable remediation options with exact commands and scoped cluster permissions. 

SREs remain in total control: once the proposal is reviewed and approved, OpenShift Lightspeed safely applies the fix in an isolated, sandboxed environment and verifies the issue is resolved. Every phase is logged for compliance, keeping operators in the driver’s seat while the agent handles execution.

These capabilities are available on Red Hat OpenShift 5.0 and later.

Red Hat OpenShift Lightspeed is available through OpenShift OperatorHub. It requires a supported edition of Red Hat OpenShift 4.16 or later, along with a supported LLM provider such as IBM watsonx.ai, Microsoft Azure OpenAI, OpenAI, Google Vertex AI, AWS Bedrock, or a private model hosted on a Red Hat-supported platform such as Red Hat Enterprise Linux® AI and Red Hat OpenShift AI. Certain features require Red Hat OpenShift 5.0 or later.