SWE-bench Verified — leaderboard
The 'Verified' subset of SWE-bench - 500 human-validated GitHub issues ensuring correct problem statements and test coverage. The most trusted variant for measuring real-world software engineering ability.
Metric: Resolved (%). Source: www.swebench.com. Status: saturation imminent. 47 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Opus 4.6 | 75.6 |
| 2 | GPT-5 (High) | 72.8 |
| 3 | GPT-5 Codex | 72.8 |
| 4 | GPT-5.2 (High) | 71.8 |
| 5 | Kimi K2.5 | 70.8 |
| 6 | Claude Sonnet 4.5 | 70.6 |
| 7 | Gemini 3 Pro | 69.6 |
| 8 | GPT-5.2 | 69 |
| 9 | Claude Opus 4 (20250514) | 67.6 |
| 10 | GPT-5.1 (Medium) | 66 |
| 11 | GPT-5 (Medium) | 65 |
| 12 | Claude Sonnet 4 (20250514) | 64.9 |
| 13 | Kimi K2 (Thinking) | 63.4 |
| 14 | MiniMax-M2 | 61 |
| 15 | GPT-5 Mini (Medium) | 59.8 |
Interactive version: theaggregate.ai/benchmark?slug=swe-bench-verified · How the rankings work · Data refreshed daily, snapshot 2026-07-22.