CASTER: Breaking the Cost-Performance Barrier in Multi-Agent Orchestration via Context-Aware Strategy for Task Efficient Routing

TL;DR

CASTER is a dynamic, context-aware routing strategy for multi-agent systems that significantly reduces inference costs while maintaining high success rates across various complex tasks.

Contribution

It introduces a novel Dual-Signal Router with self-optimizing training, enabling efficient model selection in graph-based MAS for diverse applications.

Findings

Reduces inference cost by up to 72.4%

Matches success rates of strong-model baselines

Outperforms heuristic routing and FrugalGPT across domains

Abstract

Graph-based Multi-Agent Systems (MAS) enable complex cyclic workflows but suffer from inefficient static model allocation, where deploying strong models uniformly wastes computation on trivial sub-tasks. We propose CASTER (Context-Aware Strategy for Task Efficient Routing), a lightweight router for dynamic model selection in graph-based MAS. CASTER employs a Dual-Signal Router that combines semantic embeddings with structural meta-features to estimate task difficulty. During training, the router self-optimizes through a Cold Start to Iterative Evolution paradigm, learning from its own routing failures via on-policy negative feedback. Experiments using LLM-as-a-Judge evaluation across Software Engineering, Data Analysis, Scientific Discovery, and Cybersecurity demonstrate that CASTER reduces inference cost by up to 72.4% compared to strong-model baselines while matching their success rates, and consistently outperforms both heuristic routing and FrugalGPT across all domains.