Unverified60% confidenceFactExact time
SWE-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
2
Sources
60%
Confidence
Long-term
Relevance
6/6/2025
First Seen
Sources
Gemini 2.5 Pro 0605实测对比o3与Claude Opus 4：编程、推理、写作全维度评测
bilibilikate人不错6/6/2025
Related Entities
Related Claims
Cite This Claim
Stable URI
https://kongchang.com/claim/1274API
curl https://kongchang.com/api/v1/knowledge/claims/1274MCP
get_claim(id=1274)