Expired50% confidenceFactExact time
The SWE-bench evaluation benchmark collects hundreds of issues from real GitHub repositories along with their corresponding fixes.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
7/2/2026
First Seen
Valid until: 9/30/2026(expired)
Sources
OpenAI Codex Deep Dive: The Leap from ChatGPT to an AI Coding Agent
bilibili饭饭大王ouo6/23/2026
Related Claims
UnverifiedSWE-bench Verified 由普林斯顿大学研究团队开发，从 GitHub 真实仓库抽取需修复的 Issue79% similarVerifiedSWE-bench requires models to locate and fix actual issues in real GitHub repositories.76% similarVerifiedSWE-bench (the real GitHub Issue fix rate) is regarded by the industry as the most valuable metric for measuring coding Agent capabilities.76% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/42128API
curl https://kongchang.com/api/v1/knowledge/claims/42128MCP
get_claim(id=42128)