Swe Bench
/All repositories
Overview
HEALTH SCORE
Unavailable
—
No scoring data is available. This project has no indexed repositories or the connected platform has no supported data pipeline.
LIFECYCLE
Active
Consistent commits, responsive maintainers, regular releases, and healthy issue triage.
Health breakdown
Maintainer availability, known risks and supply chain posture, and whether the project is actively developed or appropriately stable.
Maintainer responsiveness
No PRs or issues were opened in the review window, so response time could not be measured.
Bus factor
No data available. Repository has not been indexed for bus factor.
Organization Diversity
Contributors span 60 organizations.
Project Achievements
Leaderboards
About
SWE-bench is a benchmark and dataset for evaluating software engineering capabilities of large language models (LLMs). It contains real-world bug fixes from GitHub pull requests, allowing assessment of LLMs' ability to understand and fix code issues in production environments.
First commit
Software value
$2.1M
Tags
benchmark
language-model
software-engineering
License(s)
MIT
Data sources
Git
GitHub