{"slug":"swe-bench","name":"SWE-bench","category":"eval","categoryName":"Evaluation, red-teaming & guardrails","audience":["ai-era"],"primary":"ai-era","openClosed":"open","pricing":"open-source","license":"MIT","maintained":"yes","officialUrl":"https://www.swebench.com","repo":"https://github.com/SWE-bench/SWE-bench","latestVersion":"v5.0.1","latestRelease":"2026-08-18","stars":6068,"starsAsOf":"2026-10-22","description":"SWE-bench: Can Language Models Resolve Real-world Github Issues?","descriptionZh":"真实 GitHub issue 修复基准","callMethods":["CLI"],"platforms":[],"addedAt":"2026-10-10","updatedAt":"2026-10-22","topics":["benchmark","language-model","software-engineering"],"language":"Python","lastPush":"2026-09-18","formerly":null,"packages":[{"registry":"pypi","name":"swebench","url":"https://pypi.org/project/swebench/","versionsUrl":"https://pypi.org/project/swebench/#history","latestVersion":"5.0.2","latestDate":"2026-08-18"}],"recentVersions":[{"version":"v5.0.1","date":"2026-08-18","prerelease":false,"url":"https://github.com/SWE-bench/SWE-bench/releases/tag/v5.0.1","source":"GitHub release/tag 页"},{"version":"v5.0.0","date":"2026-08-17","prerelease":false,"url":"https://github.com/SWE-bench/SWE-bench/releases/tag/v5.0.0","source":"GitHub release/tag 页"},{"version":"v4.1.0","date":"2025-09-11","prerelease":false,"url":"https://github.com/SWE-bench/SWE-bench/releases/tag/v4.1.0","source":"GitHub release/tag 页"},{"version":"v4.0.4","date":"2025-07-17","prerelease":false,"url":"https://github.com/SWE-bench/SWE-bench/releases/tag/v4.0.4","source":"GitHub release/tag 页"},{"version":"v4.0.1","date":"2025-05-07","prerelease":false,"url":"https://github.com/SWE-bench/SWE-bench/releases/tag/v4.0.1","source":"GitHub release/tag 页"}],"alternatives":["osworld","lm-evaluation-harness","opencompass","tau2-bench","lmms-eval","promptfoo","openai-evals","deepeval"],"stacks":[],"scenarios":["evaluate-and-monitor"],"problems":[],"reasonZh":"编程 agent 基准；Anthropic 靠改工具描述降错","sources":["profile","priority","oss","github-pool"],"record":"https://aimap.page/api/v1/software/swe-bench.json","pages":{"en":"https://aimap.page/en/software/swe-bench/","zh-hans":"https://aimap.page/zh-hans/software/swe-bench/"}}