Get started
01 / 02Router ChallengeThree illustrative scenarios

Does handing the choice to a router beat naming a model?

A router executes a routing policy per request. Vitna evaluates whether that policy actually beats a fixed model on your workload. Because routers have no published list price, the economics can only come from your own deployment telemetry and evals. The scenarios below are illustrative: no router was run for them, and they name no product.

02 / 02Illustrative evaluationCustomer support triage · decide() arithmetic

Illustrative. No router was run: the counts, fees, success rates and latencies are written to show the arithmetic, and the scenarios name no product.

Workload: illustrative/support-triage
Quality floor: 88.0%
Economic hurdle: 15.0% savings required
Veto cases: pii_redaction, escalation_detection
Router fee: $0.0000 / req
SwitchIllustrative scenario

A zero-fee router beats incumbent Model A by 51.1% on cost per successful unit while meeting the 88.0% quality floor.

CheckRequired thresholdObservedResult
quality floor88.0%92.0%Passed
critical failures0 veto failures0 failuresPassed
economic threshold15.0%+51.1%Passed
observation threshold20 units500 unitsPassed
MetricIncumbentCandidate routerDifference
Raw unit cost$0.0300$0.0150+50.0%
Task success rate90.0%92.0%+2.0%
Successful unit cost$0.0333$0.0163+51.1%
Latency p901200ms1400ms+200ms
Platform fee drag: Router platform surcharge of $0.0000 accounted for 0% of total candidate spend across 500 units.
Epistemic provenance
A router has no listed price. Effective unit cost is derived from observed token counts and routing surcharges over confirmed workload observations.
Here nothing was observed. Every count, fee and rate above is illustrative, so there is no run to cite.
Evidence sample: 500 illustrative tickets | Cost basis: retrospective_deployment_observation