Calibration
SWE-bench Pro → FrontierCode 1.1 Extended
FrontierCode 1.1 Extended is estimated from SWE-bench Pro with a Michaelis–Menten curve fitted on 6 models measured on both: y = 0.8660·x / (0.32783 + x), R² = 0.60, cross-validated error 3.0 pp. It is used for 35 estimates.
| Estimated model | SWE-bench Pro | FrontierCode 1.1 Extended | Source |
|---|---|---|---|
| Atria Dawn Preview | 59.6% | 55.9% | estimated ± 3.0 pp, low confidence |
| Claude Mythos 5 | 80.3% | 61.5% | estimated ± 3.0 pp, medium confidence |
| Claude Opus 4.5 | 57.1% | 55.0% | estimated ± 3.0 pp, low confidence |
| Claude Opus 4.6 | 53.4% | 53.7% | estimated ± 3.0 pp, low confidence |
| Claude Opus 4.7 (Adaptive) | 64.3% | 57.4% | estimated ± 3.0 pp, medium confidence |
| dots3-note Preview | 61.0% | 56.3% | estimated ± 3.0 pp, low confidence |
| GLM-5 | 55.1% | 54.3% | estimated ± 3.0 pp, low confidence |
| GPT-5.2 | 55.6% | 54.5% | estimated ± 3.0 pp, low confidence |
| GPT-5.4 | 57.7% | 55.2% | estimated ± 3.0 pp, low confidence |
| Granite 4.2 30B | 33.3% | 43.6% | estimated ± 3.0 pp, low confidence |
| Granite 4.2 8B | 19.1% | 31.9% | estimated ± 3.0 pp, low confidence |
| Hy4 preview | 65.7% | 57.8% | estimated ± 3.0 pp, medium confidence |
| Kimi K2.5 | 50.7% | 52.6% | estimated ± 3.0 pp, low confidence |
| Laguna S 2.1 | 59.4% | 55.8% | estimated ± 3.0 pp, low confidence |
| Laguna XS 2.1 | 47.6% | 51.3% | estimated ± 3.0 pp, low confidence |
| LLaDA2.2-flash | 30.1% | 41.5% | estimated ± 3.0 pp, low confidence |
| LongCat-Flash-Lite-Sparse | 40.6% | 47.9% | estimated ± 3.0 pp, low confidence |
| MAI-Thinking-1 | 52.8% | 53.4% | estimated ± 3.0 pp, low confidence |
| MiniCPM5-2B | 14.4% | 26.4% | estimated ± 3.0 pp, low confidence |
| Muse Glimmer 30B | 51.2% | 52.8% | estimated ± 3.0 pp, low confidence |
| Ornith-1.0-35B | 50.4% | 52.5% | estimated ± 3.0 pp, low confidence |
| Ornith-1.0-397B | 62.2% | 56.7% | estimated ± 3.0 pp, low confidence |
| Ornith-1.0-9B | 42.9% | 49.1% | estimated ± 3.0 pp, low confidence |
| Ornith-1.5-35B-A3B | 59.6% | 55.9% | estimated ± 3.0 pp, low confidence |
| Ornith-1.5-397B | 65.1% | 57.6% | estimated ± 3.0 pp, medium confidence |
| Ornith-1.5-9B | 47.5% | 51.2% | estimated ± 3.0 pp, low confidence |
| Qwen3.5 397B | 50.9% | 52.7% | estimated ± 3.0 pp, low confidence |
| Qwen3.6-35B-A3B | 49.5% | 52.1% | estimated ± 3.0 pp, low confidence |
| Qwen3.7 Plus | 57.6% | 55.2% | estimated ± 3.0 pp, low confidence |
| Qwen3.8-Flash-Next | 62.5% | 56.8% | estimated ± 3.0 pp, low confidence |
| Qwen3.8-Omni-Flash | 63.3% | 57.1% | estimated ± 3.0 pp, medium confidence |
| Beam | 65.5% | 57.7% | estimated ± 3.0 pp, medium confidence |
| Sakana Fugu | 59.0% | 55.7% | estimated ± 3.0 pp, low confidence |
| Sakana Fugu-Ultra | 73.7% | 59.9% | estimated ± 3.0 pp, medium confidence |
| Step 3.7 Flash | 56.3% | 54.7% | estimated ± 3.0 pp, low confidence |