Vendor-neutral decision instrument

Route by evidence,
not reputation.

Compare fictional profiles against your actual constraints, expose the arithmetic, and leave with a policy your team can inspect.

Build policy

Define the route

Every value below is an editable assumption. No live model data or pricing is used.

Illustrative local data

Workload assumptions

Interactive support · balanced reasoning

Task archetype Privacy requirement Latency target Reasoning depth Required context Reliability target Evaluation runs Monthly requests Monthly budget assumption ($, illustrative)

Recommended route

PRIMARY

Swiftline

Primary

→

Fallback

→

TerminalHuman review

Illustrative monthly estimate

Evaluation set estimate

Dimension weights

Auditable ranking

ProfileEligibilityScoreWhy

Inspect the fictional profiles

Edit the local teaching data. Changes flow into ranking, cost, policy, and tests.

Make failure behavior explicit

A route is incomplete until timeout, uncertainty, policy, and overflow have destinations.

Primary override Secondary route Terminal behavior

Fallback triggersTimeoutLow confidencePolicy mismatchContext overflowTransient error

One endpoint is only useful when the decision behind it stays visible.

A credible router records what the task needs, which candidates failed hard constraints, how every score was assembled, what the fallback will do, what the evaluation costs under stated assumptions, and which tests must pass before production traffic moves.

Evaluation test matrix

Expected values are generated from the current route. Observations and evidence remain yours.

Test	Threshold	Expected route	Observed	Status	Evidence

Take the decision, assumptions included.

The export preserves the formula version, profile snapshot, rejected constraints, fallback triggers, matrix, and evidence.

Audit log

Model Routing Decision Logic

Read the explanation

Routecraft evaluates candidate model profiles through hard constraints and weighted scoring to pick an auditable primary and secondary path. Adjusting the latency slider down filters out slower models instantly. Only eligible candidates enter the weighted scoring formula. Explicit fallback triggers cascade execution from Swiftline to Deepframe or terminal human review upon timeout or policy mismatch.