SWE-bench Verified
Code · 2024-08-13
SWE-bench Verified is a human-validated subset of 500 SWE-bench tasks drawn from real GitHub issues in Python repositories. It measures the percentage of issues resolved by a model-generated patch that passes the repository tests.
Top models (higher is better)
| Model | Score |
|---|---|
| Claude Opus 5 | 97.0 |
| GPT-5.6 Sol | 96.2 |
| Claude Mythos 5 | 95.5 |
| Claude Fable 5 | 95.0 |
| Claude Mythos Preview | 93.9 |
| Kimi K3 | 93.4 |
| GPT-5.6 Luna | 93.0 |
| Opus 4.8 | 88.6 |
| Grok 4.5 | 86.6 |
| Macaron-V1-Venti | 85.6 |
| Sonnet 5 | 85.2 |
| GPT-5.5 | 82.9 |
| GLM-5.2 | 82.8 |
| Opus 4.7 | 82.0 |
| Muse Spark 1.1 | 82.0 |
| Opus 4.6 | 80.8 |
| Opus 4.5 | 80.6 |
| Gemini 3.1 Pro Preview | 80.6 |
| MiniMax M3 | 80.5 |
| Inkling-Small | 80.2 |
| GPT-5.2 | 80.0 |
| Sonnet 4.6 | 79.6 |
| Composer 2.5 | 79.6 |
| Gemini 3.6 Flash | 79.6 |
| Gemini 3.5 Flash | 79.3 |
| DeepSeek-V4-Flash | 79.0 |
| GPT-5.4 | 78.2 |
| Kimi K2.7 Code | 78.2 |
| GPT-5.3-Codex | 78.0 |
| DeepSeek-V4-Pro | 77.6 |
| Inkling | 77.6 |
| Inkling-Small Preview | 77.4 |
| Sonnet 4.5 | 77.2 |
| Kimi K2.6 | 76.7 |
| Qwen3.6-Max-Preview | 76.7 |
| GLM-5.1 | 76.4 |
| Gemini 3 Pro Preview | 76.4 |
| Qwen3.5 397B A17B | 76.4 |
| GPT-5.6 Terra | 75.2 |
| Gemini 3.5 Flash-Lite | 75.0 |