Playground

Route-only classification and a full test chat.

Route-only classification and a full test chat.

This screen belongs to the Router deployment operator, not an ordinary hosted customer. Hosted customers have a separate account page with their own keys and limits; policy and service settings are not exposed there. Contact the operator to change protection. In the operator console, open the route listed above; a report tenant filter does not automatically scope a global setting.

Examples use educational data. Use your own authorized tenant in real integrations. Each screenshot shows the actual frontend, with the source and educational/demo mode stated below it.

Step-by-step actions

  1. Open Playground → Route, enter synthetic text and leave model empty.
  2. Click Route and inspect sensitivity, classes, injection, action and reason.
  3. Switch to Chat, select an available model and send clean synthetic text.
  4. Stop generation if needed, then locate the request ID in history.
  5. Repeat positive and negative samples after policy changes.

Every control and its meaning

Control / parameterValues, units and effect
Маршрут / ЧатRoute classifies and selects perimeter without an LLM call. Chat calls the model and can spend budget.
Промпт / PromptSynthetic text or a preset. Do not use real secrets when testing external modes.
Модель / ModelEmpty means automatic selection; an explicit model must be available in Providers and compatible with policy.
Отправить / Stop / New chatCtrl+Enter sends the message; Stop cancels the current generation; New chat clears the local conversation.

Verify the result

Classification, route selection, model response and audit entry are separate results; verify each required step.

Errors, disabling and reverting changes

For 413, shorten the prompt or review its cap; for 429/402, check rate/budget; investigate model errors under Providers. New chat does not remove server audit.

Dry-run: PUBLIC → route → external
Dry-run: PUBLIC → route → external
2026-10-15 · Actual staging application. Synthetic client, public definitions or aggregate test-stand metrics only; not customer statistics or a benchmark. Original pixels are unmodified; personal profile and private data excluded. Dry-run checks route selection and does not call the LLM.
  1. Route
  2. Chat
  3. Harmless
  4. Code
  5. Confidential
  6. Restricted
  7. Model for route dry-run
  8. Route

Detailed security settings · Traffic analysis · FAQ · All application sections