Models

Audited model usage, cost and latency comparison.

Audited model usage, cost and latency comparison.

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 Models and refresh.
  2. Check source and tenant, then locate the model from the investigated request.
  3. Compare input/output, cost, P95 and errors with its provider.
  4. Test an alternative in Playground before changing the application or policy.

Every control and its meaning

Control / parameterValues, units and effect
Модель / провайдер / периметрLLM usage from audit; unused models may be absent. Configured models are listed under Providers.
Requests, Input/Output, costRecorded requests, tokens and USD; cost depends on configured prices and available usage.
P50/P95, classify, errorsLatency percentiles and upstream errors differ from classification time. They do not configure timeouts.
Альтернатива / AlternativeA compatibility/cost recommendation, not an automatic switch. Check capabilities and perimeter before changing models.

Verify the result

The new request must use the intended model, allowed perimeter and required capabilities. Estimated savings do not guarantee the provider tariff.

Errors, disabling and reverting changes

An empty audit or blocks-only history can produce no model rows. Check an allowed synthetic call and storage. Revert model selection in the app/policy, not this table.

Model summary · test-staging aggregates
Model summary · test-staging aggregates
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. Shown costs and latency concern test traffic; they are not pricing or an SLA.
  1. Refresh
  2. Model metrics

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