SWE-Bench++: A Framework for the Scalable Generation of Software Engineering Benchmarks from Open-Source Repositories

TL;DR

SWE-Bench++ is an automated, scalable framework that generates multilingual, repository-level software engineering benchmarks from open-source GitHub projects, enabling better evaluation and training of large language models.

Contribution

It introduces an automated pipeline for creating dynamic, repository-level coding tasks from live pull requests across multiple languages, surpassing prior manual and static datasets.

Findings

Initial benchmark has 11,133 instances from 3,971 repositories.

State-of-the-art models achieve pass@10 scores around 36%.

Fine-tuning on SWE-Bench++ improves performance on existing benchmarks.

Abstract

Benchmarks like SWE-bench have standardized the evaluation of Large Language Models (LLMs) on repository-level software engineering tasks. However, these efforts remain limited by manual curation, static datasets, and a focus on Python-based bug fixes. We introduce SWE-Bench++, an automated framework that generates repository-level coding tasks from open-source GitHub projects. Unlike synthetic approaches, our pipeline harvests live pull requests to cover both bug fixes and feature requests across 11 languages. SWE-Bench++ turns GitHub pull requests (PRs) into reproducible, execution-based tasks via four stages: programmatic sourcing, environment synthesis, test oracle extraction, and quality assurance. A final hint-guided trajectory synthesis step converts instances that strong models fail on into training trajectories. Our initial benchmark consists of 11,133 instances from 3,971 repositories across 11 languages. On a subset of 1,782 instances of this benchmark, today's strongest models perform as follows: claude-sonnet-4.5 achieves 36.20% pass@10, gpt-5-2025-08-07 34.57%, gemini/gemini-2.5-pro 24.92%, and gpt-4o 16.89%. We further demonstrate the utility of our dataset by showing that fine-tuning on SWE-Bench++ instances yields measurable improvements on the SWE-bench Multilingual benchmark. SWE-Bench++ provides a scalable, multilingual benchmark for evaluating and improving repository-level code generation.