Unverified50% confidenceFactExact time
SWE-Bench由普林斯顿大学和芝加哥大学联合推出，衡量模型解决真实GitHub Issue的能力
1
Sources
50%
Confidence
Long-term
Relevance
7/11/2026
First Seen
Sources
GPT Live全双工、Grok 4.5编程、Seedream 5.0图像：AI三赛道同步提速
bilibili养老院killer7/9/2026
Related Claims
UnverifiedSWE-bench是由Princeton和Chicago大学研究者联合构建的基准测试，包含来自真实GitHub仓库的2294个Issue修复任务82% similarPartially VerifiedSWE-Bench was developed by a Princeton University team and extracts tasks from real GitHub issues, requiring models to locate problems within complete codebases and generate fix patches78% similarUnverifiedSWE-Bench由普林斯顿大学研究团队发布，从GitHub上收集真实的Issue报告和对应的Pull Request修复补丁77% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/488374API
curl https://kongchang.com/api/v1/knowledge/claims/488374MCP
get_claim(id=488374)