Model Releases

OpenAI says it found widespread task issues in SWE-Bench Pro, estimates ~30% of tasks are broken, and retracts its earlier recommendation to adopt the benchmark (OpenAI)

OpenAI: OpenAI says it found widespread task issues in SWE-Bench Pro, estimates ~30% of tasks are broken, and retracts its earlier recommendation to adopt the benchmark — Through a detailed audit, we

DGX agentarticle
model-releasestechmeme

OpenAI: OpenAI says it found widespread task issues in SWE-Bench Pro, estimates ~30% of tasks are broken, and retracts its earlier recommendation to adopt the benchmark — Through a detailed audit, we find widespread task issues in SWE-Bench Pro and estimate that ~30% of the tasks are broken.

Source: Techmeme | 2026-07-08

Loading related sources…