SWE Atlas - Codebase QnA
Code · 2026-03-04
Score on SWE Atlas codebase QnA tasks that require runtime analysis and multi-file reasoning.
Top models (higher is better)
| Model | Score |
|---|---|
| Opus 4.8 | 57.3 |
| GLM-5.2 | 48.1 |
| Laguna S 2.1 | 46.2 |
| GPT-5.6 Sol | 46.0 |
| GPT-5.5 | 45.4 |
| Muse Spark 1.1 | 42.2 |
| GPT-5.4 | 40.8 |
| Opus 4.7 | 40.3 |
| Claude Fable 5 | 39.0 |
| Opus 4.6 | 33.3 |
| GPT-5.3-Codex | 32.6 |
| Sonnet 4.6 | 31.2 |
| DeepSeek-V4-Pro | 27.2 |
| Muse Spark | 24.2 |
| GLM-5 | 20.5 |
| Gemini 3.1 Pro Preview | 13.5 |
| Kimi K2.5 | 13.1 |
| MiniMax M2.5 | 10.3 |
| Gemini 3 Flash Preview | 8.2 |