Calibration
SWE-bench Verified → AA Coding Index
AA Coding Index is estimated from SWE-bench Verified with a offset logistic curve fitted on 32 models measured on both: y = 0.2193 + (0.8352 − 0.2193) / (1 + exp(−12.93·(x − 0.7615))), R² = 0.91, cross-validated error 5.1 pp. It is used for 33 estimates.
| Estimated model | SWE-bench Verified | AA Coding Index | Source |
|---|---|---|---|
| BTL-4 | 78.4% | 57.2% | estimated ± 5.1 pp, medium confidence |
| Claude 3.5 Sonnet | 49.0% | 23.7% | estimated ± 5.1 pp, medium confidence |
| Claude 4.1 Opus | 74.5% | 49.4% | estimated ± 5.1 pp, medium confidence |
| Claude 4 Sonnet | 72.7% | 46.0% | estimated ± 5.1 pp, medium confidence |
| Claude Haiku 4.5 | 73.3% | 47.1% | estimated ± 5.1 pp, medium confidence |
| Claude Mythos 5 | 95.5% | 78.9% | estimated ± 5.1 pp, medium confidence |
| Claude Sonnet 4.5 | 77.2% | 54.8% | estimated ± 5.1 pp, medium confidence |
| GLM-5 | 77.8% | 56.0% | estimated ± 5.1 pp, medium confidence |
| GPT-4.1 | 54.6% | 25.5% | estimated ± 5.1 pp, medium confidence |
| GPT-5.2 | 80.0% | 60.2% | estimated ± 5.1 pp, medium confidence |
| GPT-5.3 Codex | 85.0% | 68.6% | estimated ± 5.1 pp, medium confidence |
| Granite 4.2 30B | 57.0% | 26.7% | estimated ± 5.1 pp, medium confidence |
| Grok 4.20 | 76.7% | 53.8% | estimated ± 5.1 pp, medium confidence |
| Grok Code Fast 1 | 70.8% | 42.5% | estimated ± 5.1 pp, medium confidence |
| K-EXAONE 2.0 | 68.2% | 38.2% | estimated ± 5.1 pp, medium confidence |
| Laguna M.1 | 74.6% | 49.6% | estimated ± 5.1 pp, medium confidence |
| Laguna XS.2 | 69.9% | 40.9% | estimated ± 5.1 pp, medium confidence |
| Laguna XS 2.1 | 70.9% | 42.7% | estimated ± 5.1 pp, medium confidence |
| LLaDA2.2-flash | 49.3% | 23.8% | estimated ± 5.1 pp, medium confidence |
| LongCat-Flash-Lite-Sparse | 68.2% | 38.2% | estimated ± 5.1 pp, medium confidence |
| MAI-Code-1.1-Flash | 72.6% | 45.8% | estimated ± 5.1 pp, medium confidence |
| MAI-Thinking-1 | 73.5% | 47.5% | estimated ± 5.1 pp, medium confidence |
| MiMo-V2-Omni | 74.8% | 50.0% | estimated ± 5.1 pp, medium confidence |
| MiMo-V2-Pro | 78.0% | 56.4% | estimated ± 5.1 pp, medium confidence |
| MiniCPM5-2B | 46.4% | 23.2% | estimated ± 5.1 pp, medium confidence |
| o3-mini | 49.3% | 23.8% | estimated ± 5.1 pp, medium confidence |
| Qwen3.5-27B | 72.4% | 45.4% | estimated ± 5.1 pp, medium confidence |
| Qwen3.5-35B-A3B | 69.2% | 39.7% | estimated ± 5.1 pp, medium confidence |
| Qwen3.5 397B | 76.2% | 52.8% | estimated ± 5.1 pp, medium confidence |
| Solar Open 2 | 70.4% | 41.8% | estimated ± 5.1 pp, medium confidence |
| Solar Pro 4 | 70.6% | 42.1% | estimated ± 5.1 pp, medium confidence |
| Ternary Bonsai 2 27B | 60.8% | 29.4% | estimated ± 5.1 pp, medium confidence |
| ZAYA1-74B-Preview | 53.2% | 24.9% | estimated ± 5.1 pp, medium confidence |