Configure STRATUM AI
Configure provider endpoint/model/key and enable STRATUM AI.
Provider and execution controls
Enable STRATUM AI, select the provider type, and enter the model, provider endpoint, and credential required by that provider. Test the connection before relying on AI operations elsewhere in the product.
The compute-control settings bound how much autonomous work an AI operation may perform. Set those limits to match the environment’s operating policy.
The selected AI model still sets the ceiling for reasoning quality and tool planning. A weaker model can produce weaker plans, but STRATUM keeps the operator in control: plans can be rejected, corrected, or refined before execution. STRATUMvisor is developing a datacenter-operations-tuned model that can be offered as a service or run on-premises, focusing the model on infrastructure operations instead of requiring broad general-world knowledge.
Interface tour
AI Provider Configuration
Enable STRATUM AI, select the provider, and enter the model, provider base URL, and credential required for that service. Test the connection before using AI operations.
AI Compute Control
Set provider timeout, token/output limits, and operation limits for this controller. Enable VM Console Computer Use only when a vision-capable model should be allowed temporary, user-granted control of one STRATUM VM console at a time. STRATUM normally operates VMs and switches through backend APIs and Guest Tools; when Guest Tools is not installed in a VM, console computer use provides a controlled fallback path so the AI can still interact with that guest through its console.
The outcome you should see
STRATUM AI is enabled with a valid provider/model configuration and compute-control limits reflect the operator policy.