A software-engineering benchmark built from real Python repositories and GitHub issues to test whether models can resolve practical coding tasks.
Browse the latest scores, model modes, release dates, and parameter sizes for SWE-bench.
Data sourced primarily from official releases (GitHub, Hugging Face, papers), then benchmark leaderboards, then third-party evaluators. Learn about our data methodology
| Rank | Model | License | |||
|---|---|---|---|---|---|
 Claude Opus 4.6 Extended ThinkingTools | 77.83 | 2026-02-05 | Unknown | Closed | |
 Grok 4 Code Standard Mode | 72.00 | 2025-07-03 | Unknown | Closed |
