Introducing SWE-bench Verified

Company news OpenAI Source language: English

We’re releasing a human-validated subset of SWE-bench that more reliably evaluates AI models’ ability to solve real-world software issues.

Excerpt

We’re releasing a human-validated subset of SWE-bench that more reliably evaluates AI models’ ability to solve real-world software issues.

Short excerpt from the source — open the original for the full text.

Sources

← Back to the daily brief