Optimizing Sequential Multi-Step Tasks with Parallel LLM Agents

TL;DR

This paper introduces M1-Parallel, a framework that runs multiple multi-agent plans concurrently to reduce latency and improve success rates in complex reasoning tasks involving large language models.

Contribution

The paper presents M1-Parallel, a novel parallel execution framework for multi-agent LLM systems, enhancing efficiency and effectiveness in complex task solving.

Findings

M1-Parallel with early termination achieves up to 2.2x speedup.

Parallel execution increases task completion rates.

Diversity strategies did not improve performance over repeated sampling.

Abstract

Large language model (LLM)-based multi-agent systems have demonstrated remarkable promise for tackling complex tasks by breaking them down into subtasks that are iteratively planned, executed, observed, and refined. Despite their effectiveness, these systems often incur high latency because real-world problems frequently demand multiple iterative cycles of reasoning steps. To address this challenge, we propose M1-Parallel, a framework that concurrently runs multiple multi-agent teams in parallel to uncover distinct solution paths. By leveraging an event-driven communication model with asynchronous messaging, M1-Parallel efficiently capitalizes on the inherent diversity of valid plans to either reduce end-to-end latency or boost task completion rates. Our experiments on complex tasks show that M1-Parallel with early termination achieves up to $2.2\times$ speedup while preserving accuracy, and that M1-Parallel with aggregation yields higher task completion rates. We further investigate strategies aimed at encouraging diverse execution plans but observe no additional performance gains over repeated sampling. Overall, these findings underscore the potential of parallel plan execution for optimizing multi-agent systems for real-world, high-complexity reasoning tasks.