SEAR: Schema-Based Evaluation and Routing for LLM Gateways
This work addresses the need for fine-grained, interpretable quality assessment and efficient routing in production LLM gateways, offering a practical solution for organizations managing multiple LLM providers.
The paper tackles the problem of evaluating LLM responses and routing requests in multi-provider gateways by introducing SEAR, a schema-based system that uses LLM reasoning to produce structured evaluation signals and operational metrics, achieving strong accuracy on human-labeled data and enabling significant cost reductions while maintaining quality.
Evaluating production LLM responses and routing requests across providers in LLM gateways requires fine-grained quality signals and operationally grounded decisions. To address this gap, we present SEAR, a schema-based evaluation and routing system for multi-model, multi-provider LLM gateways. SEAR defines an extensible relational schema covering both LLM evaluation signals (context, intent, response characteristics, issue attribution, and quality scores) and gateway operational metrics (latency, cost, throughput), with cross-table consistency links across around one hundred typed, SQL-queryable columns. To populate the evaluation signals reliably, SEAR proposes self-contained signal instructions, in-schema reasoning, and multi-stage generation that produces database-ready structured outputs. Because signals are derived through LLM reasoning rather than shallow classifiers, SEAR captures complex request semantics, enables human-interpretable routing explanations, and unifies evaluation and routing in a single query layer. Across thousands of production sessions, SEAR achieves strong signal accuracy on human-labeled data and supports practical routing decisions, including large cost reductions with comparable quality.