Architecture Advanced 2 minute read Updated 2026-06-26 UTC

Router and coalition selection

How to select specialists, cascades, and small ensembles under capability, confidence, cost, and safety constraints.

Research statusEstablished orchestration patterns Publication statePublished Reviewed byMichael Kappel Source reports3

Routing is a policy decision

The router chooses which model or coalition receives a task. It affects quality, cost, data exposure, and which specialists receive evidence. Treat it as a governed component with its own lineage and evaluation—not as an invisible convenience layer.

Routing strategies

StrategyStrengthLimitationGood use
Static rulesPredictable and auditableBrittle as tasks changeClear domains and compliance boundaries
Capability lookupSimple modularityDepends on accurate metadataContract-driven systems
Learned classifierAdapts to complex inputsCan drift or bias trafficHigh-volume stable taxonomies
Cost-aware optimizerExplicit budget trade-offsRequires reliable profilesMulti-tier edge/cloud systems
CascadeSaves average costHigh worst-case latencyEasy versus hard query separation
Parallel coalitionImproves robustnessExpensive and correlatedHigh-value uncertain tasks
Champion–challenger splitProduces comparison evidenceMust protect users from challengersShadow and canary evaluation

Eligibility before ranking

First filter by hard predicates: capability contract, data jurisdiction, risk tier, runtime compatibility, permissions, current health, and resource ceiling. Ranking only happens among eligible models.

pseudocode
FUNCTION route(request, population, ledger)
    eligible <- []

    FOR each model IN population
        IF contract_matches(model, request)
           AND policy_allows(model, request)
           AND resource_profile_fits(model, ledger)
           AND health_is_acceptable(model)
            APPEND eligible, model
        END IF
    END FOR

    IF eligible IS EMPTY
        RETURN approved_fallback_plan(request)
    END IF

    ranked <- SCORE_BY_EXPECTED_VALUE(eligible, request, ledger)
    RETURN BUILD_BOUNDED_PLAN(ranked, request.risk_tier)
END FUNCTION

Coalition rules

Cap coalition size. Prefer independent generation followed by a fixed aggregator or judge. Define timeouts, quorum, disagreement behavior, and maximum total cost. Do not let models recruit additional models dynamically unless the planner itself is governed and budgeted.

Exploration without user harm

Use shadow routing to collect challenger predictions without affecting responses. For canaries, restrict traffic by risk tier, cohort, geography, or request type. Exploration quotas should be explicit and reversible.

Router feedback loops

A router can starve a specialist of traffic, then conclude it lacks evidence or quality. It can also create popularity loops where already-selected models receive more updates and become increasingly dominant. Preserve evaluation traffic, use counterfactual or offline datasets, and measure traffic concentration.

Router metrics

Track routing accuracy, fallback rate, abstention, cost per accepted result, p95/p99 latency, specialist utilization, traffic entropy, disagreement rate, and quality by route. Evaluate the router and model population jointly because a strong router can hide weak models and vice versa.

Source reports used for this guide

These reports are preserved verbatim in the site archive. The guide above is an editorial synthesis and may narrow, qualify, or reorganize claims from the source material.