Topic 7 of 20 · Part II : Grounding, Evaluation and Orchestration

Model routing

Routing is not merely cost optimisation : it improves quality by sending each request to the component best suited to it.

The control pattern

Data asset

Request characteristics and runtime signals

Knobs

Complexity, Domain, Risk, Confidence, Latency and cost budgets, Tool requirements

Outcome

The right model and workflow for each request

Measurement

See how to measure this below

What it is

Model routing avoids using the same model for every task. The incoming request and its metadata are the data asset, while complexity, risk, domain, confidence and service-level requirements are the routing knobs.

A lightweight classifier can recognise the task and select a deterministic function, a retrieval workflow, a distilled student, a larger reasoning model, or human escalation. Routing is not merely cost optimisation. It improves quality by sending structured calculations to code, current-price questions to a market-data tool, and nuanced cross-document analysis to a model capable of deeper reasoning.

A strong router is also what makes distillation practical: the student handles the high-volume centre of the distribution while the teacher remains available for the difficult tail.

The knobs in detail

Each row is one adjustable property of the data asset, and what moving it tends to do.

KnobWhat you adjustLikely effect
ComplexityHow many documents and steps are involvedSelects between fast and deep paths
DomainWhat kind of question is being askedRoutes to the right specialist or tool
RiskConsequence of being wrongRaises evidence and explanation requirements
ConfidenceThe student's own certaintyTriggers escalation to a teacher model
Latency and cost budgetsService-level targets per surfaceKeeps interactive paths fast
Tool requirementsWhether code or a live feed is neededSends calculations to deterministic components

Applied: Stocks Assistant

For Stocks Assistant, simple ticker lookup, schema validation and standard earnings extraction can use fast specialised components. A question comparing several companies across four quarters may require richer retrieval and a larger model. A request combining options exposure, portfolio concentration, taxes and personalised action carries higher risk and should face stricter evidence and explanation requirements. The router should also detect missing context: “Should I sell?” cannot be answered responsibly without horizon, objective, position context and an explicit description of uncertainty.

How to measure it

Evidence that the knob produced the intended behaviour, rather than shifting the problem elsewhere.

  • Routing accuracy, evaluated separately from answer quality
  • Logged intent, chosen route, confidence, tools called and fallback reason
  • Latency and cost per route
  • End-to-end answer quality by route

Common mistakes

Evaluating only the downstream answer. A perfect model cannot rescue a misclassified request.

Unversioned routing rules, which make regressions impossible to attribute.