Trae Agent: An LLM-based Agent for Software Engineering with Test-time Scaling

2025-07-31

AI Summary

Existing LLM prompting methods struggle to efficiently explore large integration search spaces and lack repository-level contextual understanding, limiting their effectiveness in software engineering problem repair. Method: This paper proposes the first LLM agent framework specifically designed for repository-level problem repair, introducing the agent paradigm to this task. It enables end-to-end integrative reasoning through three coordinated modules—generation, pruning, and selection—and supports dynamic test-time expansion. Contribution/Results: The design overcomes dual limitations of conventional prompting approaches: scalability in search space exploration and capacity for global semantic modeling. Evaluated on the SWE-bench benchmark, our method achieves an average 10.22% improvement in Pass@1 and attains a score of 75.20%, ranking first on the Verified leaderboard. The implementation is publicly available.

Technology Category

• Planning, Routing, and Scheduling: Replanning and Plan Repair
• Search and Optimization: Sampling/Simulation-based Search
• Multiagent Systems: Distributed Problem Solving

Application Category

• Search and Retrieval-Augmented AI: Agentic search
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation

Abstract

Software issue resolution is a critical challenge in software engineering and has garnered increasing attention in recent years. With the rapid advancement of large language models (LLMs), substantial progress has been made in addressing real-world software engineering tasks. Recent studies have introduced ensemble reasoning techniques to enhance the performance of LLM-based issue resolution. However, existing prompting-based methods still face limitations in effectively exploring large ensemble spaces and lack the capacity for repository-level understanding, both of which constrain their overall effectiveness. In this paper, we propose Trae Agent, the first agent-based ensemble reasoning approach for repository-level issue resolution. Trae Agent formulates our goal as an optimal solution search problem and addresses two key challenges, i.e., large ensemble spaces and repository-level understanding, through modular agents for generation, pruning, and selection. We conduct extensive experiments using three leading LLMs on the widely-adopted SWE-bench benchmark, comparing Trae Agent against four state-of-the-art ensemble reasoning techniques. Experimental results demonstrate that Trae Agent consistently achieves superior performance, with an average improvement of 10.22% over all baselines in terms of Pass@1. Trae Agent has achieved first place on the SWE-bench Verified leaderboard, with a notable Pass@1 score of 75.20%. We are pleased to release Trae Agent as an open-source project to support the research community, with all resources available at https://github.com/bytedance/trae-agent.

Problem

• Enhancing LLM-based software issue resolution performance
• Addressing large ensemble spaces in reasoning methods
• Achieving repository-level understanding in issue resolution

Innovation

• LLM-based agent for software issue resolution
• Modular agents handle generation, pruning, selection
• Optimal solution search with ensemble reasoning