Verified65% confidenceFactTime unknown
Devin在SWE-bench测试中解决了13.86%的真实GitHub Issues，高于Claude 2的表现
3
Sources
65%
Confidence
Long-term
Relevance
5/31/2026
First Seen
Sources
AI编程工具深度对比：Qoder、Cursor、Windsurf、Devin怎么选
bilibili烟岚九境
Related Entities
Related Claims
Unverified2023年普林斯顿大学的SWE-bench基准测试显示，当时最先进的AI编程工具在真实GitHub Issue修复任务中的成功率仅约13%73% similarUnverifiedDevin 2.0在SWE-bench基准测试中解决了13.86%的实际编程问题，而此前的AI模型仅为1.96%72% similarUnverifiedDevin在SWE-bench评测中达到13.86%的解决率（当时最高）71% similarUnverified在SWE-Bench编程基准测试中，Claude Code得分80.8%，GitHub Copilot仅为72.5%70% similarUnverifiedGitHub无障碍智能体测试中已审查3535个PR，问题解决率达到68%68% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/3203API
curl https://kongchang.com/api/v1/knowledge/claims/3203MCP
get_claim(id=3203)