Can Small Agents Collaborate to Beat a Single Large Language Model?

TL;DR

This paper demonstrates that small, well-organized multi-agent systems with an orchestrator can outperform larger single models on complex reasoning tasks, emphasizing architecture over size.

Contribution

It shows that multi-agent collaboration with a focus on orchestration can surpass larger models, challenging the reliance on scaling for performance improvements.

Findings

Small multi-agent systems outperform larger single models on reasoning tasks.

Orchestrator capacity is more critical than sub-agent capacity for performance.

Reasoning in the orchestrator yields the largest performance gains.

Abstract

Recent progress in language modeling has largely relied on scaling model size, yet larger models do not reliably improve performance on tasks requiring multi-step reasoning and tool use. Multi-agent collaboration offers a potential alternative, raising a key question: can well-organized systems built from smaller models outperform much larger language models? We address this question using a minimally designed multi-agent system with a single orchestrator and a small set of specialized sub-agents with restricted communication. On tool-intensive benchmarks spanning factual retrieval, multi-hop reasoning, scientific question answering, and mathematical problem solving, we conduct controlled comparisons between small multi-agent systems and large single-agent models. We find that small multi-agent systems can outperform substantially larger single-agent models, even when the latter have direct access to tools. Reasoning at the orchestrator yields the largest gains, while enabling reasoning in sub-agents provides limited or negative benefits. Overall system performance is driven primarily by orchestrator capacity rather than sub-agent capacity. These results suggest that improved agentic performance depends more on architectural orchestration than on raw model scaling.