SWE-Bench ProMax: Benchmarking Agents on Large-Scale Multilingual Code Refactoring
2026-08-10 • Computation and Language
Computation and LanguageSoftware Engineering
AI summaryⓘ
The authors point out that current tests for AI coding tools are flawed because many have bad or unclear test cases and models often copy known solutions. They created SWE-Bench ProMax, a new set of tough code refactoring tasks from real software projects in seven programming languages, carefully checked to fix these problems. These tasks require changing many files while keeping the program working right, making them more realistic and harder. When they tested the best AI models, none solved more than 41% of the tasks, showing this new benchmark is still challenging for AI.
AI coding agentssoftware engineering benchmarkcode refactoringtest suitemultilingual programmingbehavior-preserving changeslong-horizon tasksmodel evaluationfrontier modelsbenchmark saturation
Authors
Yuling Shi, Jinghan Xu, Kelin Fu, Wenhao Zeng, Shilin He, Lei Zhang, Yue Liu, Zelin Zhao, Terry Yue Zhuo, Jialun Cao, Siyu Ye, Tianyu Liu, Kai Cai, Shing-Chi Cheung, Xiaodong Gu
Abstract
As AI coding agents take on increasingly complex, long-horizon software engineering tasks, existing benchmarks are rapidly saturating and their evaluation quality has come under serious scrutiny: a recent audit found that nearly 60% of unsolved SWE-bench Verified instances contain flawed tests -- either overly narrow tests that reject correct solutions or overly broad tests that check unstated requirements -- and that frontier models can verbatim reproduce gold patches from training data. Code refactoring, which requires coordinated, behavior-preserving changes across many files, offers a substantially harder and more realistic test of agent capability, yet remains underserved by current benchmarks. We introduce SWE-Bench ProMax, an expert-curated, multilingual code refactoring benchmark of 170 instances drawn from real commits across seven programming languages (Python, Java, TypeScript, Go, C, C++, and Rust). Every instance undergoes rigorous, multi-stage curation that directly addresses the quality problems identified in prior benchmarks: issue descriptions are rewritten from scratch to provide precise, unambiguous specifications, and test suites are manually reviewed to remove overly narrow and overly broad tests. Tasks with insufficient complexity or limited cross-file scope are filtered out, yielding a benchmark of challenging, large-scale refactoring tasks that average 11.4 modified files and 261.6 lines of code per instance, substantially exceeding the scale of existing benchmarks. Experiments with frontier models under two agent scaffolds show that the best model achieves only 41.2% resolve rate, confirming that SWE-Bench ProMax presents a meaningful and unsaturated challenge for current AI coding agents. Our benchmark is available at https://huggingface.co/datasets/swe-bench-promax/SWE-Bench-ProMax.