SWE-Compass: Towards Unified Evaluation of Agentic Coding Abilities for Large Language Models

TL;DR

SWE-Compass is a comprehensive, multi-dimensional benchmark designed to evaluate large language models' coding abilities across diverse tasks, languages, and real-world developer workflows, addressing limitations of previous narrow and biased assessments.

Contribution

The paper introduces SWE-Compass, a unified, structured benchmark covering multiple coding tasks, scenarios, and languages, aligned with real-world software engineering practices.

Findings

Hierarchical difficulty across tasks, languages, and scenarios identified.

State-of-the-art LLMs evaluated, revealing performance gaps.

Benchmark facilitates diagnosis and advancement of agentic coding abilities.

Abstract

Evaluating large language models (LLMs) for software engineering has been limited by narrow task coverage, language bias, and insufficient alignment with real-world developer workflows. Existing benchmarks often focus on algorithmic problems or Python-centric bug fixing, leaving critical dimensions of software engineering underexplored. To address these gaps, we introduce SWE-Compass1, a comprehensive benchmark that unifies heterogeneous code-related evaluations into a structured and production-aligned framework. SWE-Compass spans 8 task types, 8 programming scenarios, and 10 programming languages, with 2000 high-quality instances curated from authentic GitHub pull requests and refined through systematic filtering and validation. We benchmark ten state-of-the-art LLMs under two agentic frameworks, SWE-Agent and Claude Code, revealing a clear hierarchy of difficulty across task types, languages, and scenarios. Moreover, by aligning evaluation with real-world developer practices, SWE-Compass provides a rigorous and reproducible foundation for diagnosing and advancing agentic coding capabilities in large language models.