Aragog: Just-in-Time Model Routing for Scalable Serving of Agentic Workflows

TL;DR

Aragog is a system that dynamically adapts model routing configurations during workflow execution, significantly improving scalability and efficiency in serving complex agentic workflows with large language models.

Contribution

It introduces a novel approach that decouples configuration identification from runtime scheduling, enabling adaptive, cost-effective workflow serving.

Findings

Increases maximum throughput by up to 217%.

Reduces median latency by up to 78.9%.

Maintains accuracy comparable to expensive configurations.

Abstract

Agentic workflows have emerged as a powerful paradigm for solving complex, multi-stage tasks, but serving them at scale is computationally expensive given the many LLM inferences that each request must pass through. Configuration selection, or the cost-aware assignment of workflow agents to specific LLMs, can reduce these costs, but existing approaches bind configuration decisions before request execution, making them ill-suited for the heterogeneous and lengthy execution of workflows. Specifically, system loads can fluctuate rapidly and substantially during a request's lifetime, causing fixed configurations to quickly become suboptimal. We present Aragog, a system that progressively adapts a request's configuration throughout its execution to match runtime dynamics. To make this practical despite the massive space of workflow configurations, Aragog decouples the problem into two core elements -- a one-time routing step that identifies all accuracy-preserving configurations, and a cheap per-stage scheduler that selects among them using up-to-date system observations -- and introduces novel strategies to accelerate each. Across diverse workflows and model families, Aragog increases maximum serving throughput by 50.0--217.0\% and reduces median latency by 32.5--78.9\% at peak request rates, while maintaining accuracy comparable to the most expensive configurations.