Expired50% confidenceBenchmarkExact time
In the SWE-bench real GitHub Issue fix task, the fix success rates of models from different vendors can differ by 15 to 30 percentage points.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
7/8/2026
First Seen
Valid until: 10/6/2026(expired)
Sources
OpenAI's Official Plugin codex-plugin-cc: Calling Codex Inside Claude Code for Dual-AI Collaboration
githubGitHub Trending7/3/2026
Related Claims
UnverifiedThe SWE-bench benchmark from Princeton University in 2023 showed that the most advanced AI programming tools at the time achieved a success rate of only about 13% on real GitHub issue fix tasks.78% similarVerifiedSWE-bench (the real GitHub Issue fix rate) is regarded by the industry as the most valuable metric for measuring coding Agent capabilities.77% similarUnverifiedSWE-bench requires models to locate and fix bugs in real GitHub repositories, with the pass rate gradually increasing from less than 2% initially to over 20%.77% similarUnverifiedSWE-bench requires models to resolve real GitHub issues, and the pass rate does not exceed 50% even for the strongest models.73% similarUnverifiedThe automatic code repair rate on the SWE-bench software engineering benchmark has risen from less than 5% initially to over 30%.71% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/308983API
curl https://kongchang.com/api/v1/knowledge/claims/308983MCP
get_claim(id=308983)