Explainable Model Routing for Agentic Workflows

TL;DR

Topaz introduces an interpretable routing framework for agentic workflows, enabling auditability, transparency, and better trade-off management between model capability and cost.

Contribution

It provides a formal, interpretable routing system with skill profiling, traceable algorithms, and natural language explanations for agentic workflows.

Findings

Enables understanding of model routing decisions.

Supports iterative tuning of cost-quality tradeoffs.

Improves trust and transparency in agentic systems.

Abstract

Modern agentic workflows decompose complex tasks into specialized subtasks and route them to diverse models to minimize cost without sacrificing quality. However, current routing architectures focus exclusively on performance optimization, leaving underlying trade-offs between model capability and cost unrecorded. Without clear rationale, developers cannot distinguish between intelligent efficiency -- using specialized models for appropriate tasks -- and latent failures caused by budget-driven model selection. We present Topaz, a framework that introduces formal auditability to agentic routing. Topaz replaces silent model assignments with an inherently interpretable router that incorporates three components: (i) skill-based profiling that synthesizes performance across diverse benchmarks into granular capability profiles (ii) fully traceable routing algorithms that utilize budget-based and multi-objective optimization to produce clear traces of how skill-match scores were weighed against costs, and (iii) developer-facing explanations that translate these traces into natural language, allowing users to audit system logic and iteratively tune the cost-quality tradeoff. By making routing decisions interpretable, Topaz enables users to understand, trust, and meaningfully steer routed agentic systems.