Calibration
SWE-bench Pro → NL2Repo
NL2Repo is estimated from SWE-bench Pro with a linear curve fitted on 28 models measured on both: y = 1.2179·x + -0.2615, R² = 0.60, cross-validated error 6.1 pp. It is used for 16 estimates.
| Estimated model | SWE-bench Pro | NL2Repo | Source |
|---|---|---|---|
| Claude Opus 5.5 | 89.9% | 83.3% | estimated ± 6.1 pp, low confidence |
| Claude Sonnet 5.5 | 81.3% | 72.9% | estimated ± 6.1 pp, low confidence |
| Claude Mythos 5 | 80.3% | 71.7% | estimated ± 6.1 pp, low confidence |
| Ling 3.1 Flash | 68.7% | 57.6% | estimated ± 6.1 pp, low confidence |
| Beam 1.0 501B-A23B | 66.1% | 54.4% | estimated ± 6.1 pp, medium confidence |
| Beam 1.0 501B-A23B FP8 | 65.9% | 54.1% | estimated ± 6.1 pp, medium confidence |
| Beam 1.0 501B-A23B NVFP4 | 65.6% | 53.7% | estimated ± 6.1 pp, medium confidence |
| Sage-A1.5 | 61.2% | 48.4% | estimated ± 6.1 pp, medium confidence |
| Atria Dawn Preview | 59.6% | 46.4% | estimated ± 6.1 pp, medium confidence |
| Laguna S 2.1 | 59.4% | 46.2% | estimated ± 6.1 pp, medium confidence |
| GPT-5.2 | 55.6% | 41.6% | estimated ± 6.1 pp, medium confidence |
| Spark-X2.6-4B | 54.3% | 40.0% | estimated ± 6.1 pp, medium confidence |
| Claude Opus 4.6 | 53.4% | 38.9% | estimated ± 6.1 pp, medium confidence |
| Laguna XS 2.1 | 47.6% | 31.8% | estimated ± 6.1 pp, medium confidence |
| LongCat-Flash-Lite-Sparse | 40.6% | 23.3% | estimated ± 6.1 pp, low confidence |
| LLaDA2.2-flash | 30.1% | 10.5% | estimated ± 6.1 pp, low confidence |
Fitted on
| Model | SWE-bench Pro | NL2Repo |
|---|---|---|
| Claude Opus 4.5 | 57.1% | 43.2% |
| DeepSeek V4 Flash 0731 | 52.6% | 54.2% |
| DeepSeek V4 Pro 0813 | 55.4% | 61.5% |
| dots3-note Preview | 61.0% | 49.8% |
| GLM-5 | 55.1% | 35.9% |
| GLM-5.1 | 58.4% | 42.7% |
| GLM-5.2 | 62.1% | 48.9% |
| Hy4 preview | 65.7% | 58.9% |
| Kimi K2.5 | 50.7% | 32.0% |
| MiniMax M2.7 | 56.2% | 39.8% |
| MiniMax M3 | 59.0% | 42.1% |
| Ornith-1.0-35B | 50.4% | 34.6% |
| Ornith-1.0-397B | 62.2% | 48.2% |
| Ornith-1.0-9B | 42.9% | 27.2% |
| Ornith-1.5-35B-A3B | 59.6% | 46.2% |
| Ornith-1.5-397B | 65.1% | 59.5% |
| Ornith-1.5-9B | 47.5% | 32.4% |
| Qwen3.5 397B | 50.9% | 32.2% |
| Qwen3.6-27B | 53.5% | 36.2% |
| Qwen3.6-35B-A3B | 49.5% | 29.4% |
| Qwen 3.6 Max (preview) | 57.3% | 42.9% |
| Qwen3.6 Plus | 56.6% | 37.9% |
| Qwen3.7 Max | 60.6% | 47.2% |
| Qwen3.7 Plus | 57.6% | 41.1% |
| Qwen3.8-27B | 61.7% | 42.3% |
| Qwen3.8-Flash-Next | 62.5% | 48.1% |
| Qwen3.8 Max | 67.7% | 55.9% |
| Qwen3.8-Omni-Flash | 63.3% | 48.9% |