SWE Pro
Can the model complete substantial programming work under this benchmark's agent setup?
Coding7 ranked models11 reported values1 reportsHigher is better
Top rankings
One row per model, using its best reported score across effort settings.
| Rank | Model | Best score | Best reported setting |
|---|---|---|---|
| 1 | Kimi K2.6Moonshot AI | 58.6% | Reported configurationDeepSeek report1 value · 1 reportApr 26, 2026 · source |
| 2 | GLM-5.1Z.ai | 58.4% | Reported configurationDeepSeek report1 value · 1 reportApr 26, 2026 · source |
| 3 | GPT-5.4OpenAI | 57.7% | xhigh effortDeepSeek report1 value · 1 reportApr 26, 2026 · source |
| 4 | Claude Opus 4.6Anthropic | 57.3% | max effortDeepSeek report1 value · 1 reportApr 26, 2026 · source |
| 5 | DeepSeek V4 ProDeepSeek | 55.4% | max effortDeepSeek report3 values · 1 reportApr 26, 2026 · source |
| 6 | Gemini 3.1 ProGoogle | 54.2% | high effortDeepSeek report1 value · 1 reportApr 26, 2026 · source |
| 7 | DeepSeek-V4-FlashDeepSeek | 52.6% | max effortDeepSeek report3 values · 1 reportApr 26, 2026 · source |
Effort curve
Every sourced cost-linked effort value for this exact version. Lines connect complete sweeps only.
No cost-linked effort sweep for this version.
Definition and comparison boundarypublic methodology
Can the model complete substantial programming work under this benchmark's agent setup?
Software implementation, debugging, or repository work under the published evaluation protocol. Higher is better. The value is the percentage reported in this lab's table.
Harness, tools, prompts, attempt count, and benchmark version can materially change the result. Compare within one reporting context.
Frontier comparison and across-modes table.
Method / source