AFlow: Automating Agentic Workflow Generation

TL;DR

AFlow is an automated framework that uses Monte Carlo Tree Search to generate and optimize agentic workflows with minimal human intervention, improving performance and reducing costs across multiple benchmarks.

Contribution

We reformulate workflow optimization as a search problem and introduce AFlow, which automates workflow generation using code modifications and feedback, outperforming existing methods.

Findings

AFlow achieves a 5.7% average improvement over baselines.

Smaller models outperform GPT-4o on certain tasks at 4.55% of its inference cost.

Empirical evaluation across six datasets demonstrates AFlow's effectiveness.

Abstract

Large language models (LLMs) have demonstrated remarkable potential in solving complex tasks across diverse domains, typically by employing agentic workflows that follow detailed instructions and operational sequences. However, constructing these workflows requires significant human effort, limiting scalability and generalizability. Recent research has sought to automate the generation and optimization of these workflows, but existing methods still rely on initial manual setup and fall short of achieving fully automated and effective workflow generation. To address this challenge, we reformulate workflow optimization as a search problem over code-represented workflows, where LLM-invoking nodes are connected by edges. We introduce AFlow, an automated framework that efficiently explores this space using Monte Carlo Tree Search, iteratively refining workflows through code modification, tree-structured experience, and execution feedback. Empirical evaluations across six benchmark datasets demonstrate AFlow's efficacy, yielding a 5.7% average improvement over state-of-the-art baselines. Furthermore, AFlow enables smaller models to outperform GPT-4o on specific tasks at 4.55% of its inference cost in dollars. The code is available at https://github.com/FoundationAgents/AFlow.