ai
Source: OpenAI News
📅 13 September 2026 at 04:31 PM UTC
Why we no longer evaluate SWE-bench Verified
Verified Primary Feed Ingestion
⚡ Executive Takeaways & Intelligence Summary
SWE-bench Verified is increasingly contaminated and mismeasures frontier coding progress. Our analysis shows flawed tests and training leakage. We recommend SWE-bench Pro.
Publication Reference: https://openai.com/index/why-we-no-longer-evaluate-swe-bench-verified
← Back to News Wire