SWE-bench Verified Epoch-Compatible 484-Task Subset
Code · 2026-02-01
The 484-task subset of SWE-bench Verified that passes on Epoch AI's evaluation infrastructure. Sixteen tasks from the 500-task parent benchmark are excluded, and individual runs may have fewer completions, so these scores are not directly comparable to full-set results.
Top models (higher is better)
| Model | Score |
|---|---|
| Opus 4.7 | 83.5 |
| GPT-5.5 | 80.6 |
| Gemini 3.5 Flash | 79.3 |
| GLM-5.2 | 78.7 |
| DeepSeek-V4-Pro | 77.6 |
| Qwen3.7-Max | 77.3 |
| GPT-5.4 | 76.9 |
| Opus 4.5 | 76.7 |
| Kimi K2.6 | 76.7 |
| Qwen3.6-Max-Preview | 76.7 |
| Gemini 3.1 Pro Preview | 75.6 |
| Gemini 3 Flash Preview | 75.4 |
| Sonnet 4.6 | 75.2 |
| GPT-5.3-Codex | 74.8 |
| GLM-5.1 | 74.2 |
| GPT-5.2 | 73.8 |
| Kimi K2.5 | 73.8 |
| GPT-5 | 73.6 |
| Opus 4.1 | 73.3 |
| Gemini 3 Pro Preview | 72.9 |
| GLM-5 | 72.1 |
| Sonnet 4.5 | 71.3 |
| Opus 4 | 70.7 |
| GPT-5 Mini | 64.7 |
| o3 | 62.3 |
| Claude 3.7 Sonnet | 61.0 |
| Qwen3.6 Plus (2026-04-02) | 57.9 |
| Gemini 2.5 Pro | 57.6 |
| GPT-4.1 | 48.5 |
| GPT-4o (2024-11-20) | 31.0 |