Calibration
SWE-bench Pro → LiveCodeBench v6
LiveCodeBench v6 is estimated from SWE-bench Pro with a linear curve fitted on 15 models measured on both: y = 0.4408·x + 0.6312, R² = 0.93, cross-validated error 2.1 pp. It is used for 24 estimates.
| Estimated model | SWE-bench Pro | LiveCodeBench v6 | Source |
|---|---|---|---|
| Atria Dawn Preview | 59.6% | 89.4% | estimated ± 2.1 pp, high confidence |
| Claude Fable 5.1 | 81.2% | 98.9% | estimated ± 2.1 pp, medium confidence |
| Claude Opus 5.5 | 89.9% | 100.0% | estimated ± 2.1 pp, medium confidence |
| Claude Sonnet 5.5 | 81.3% | 99.0% | estimated ± 2.1 pp, medium confidence |
| Gemini 3.5 Flash | 55.1% | 87.4% | estimated ± 2.1 pp, high confidence |
| Gemini 3.5 Flash-Lite | 54.2% | 87.0% | estimated ± 2.1 pp, high confidence |
| GLM-5.1 | 58.4% | 88.9% | estimated ± 2.1 pp, high confidence |
| GLM-5.2 | 62.1% | 90.5% | estimated ± 2.1 pp, high confidence |
| GPT-5.4 | 57.7% | 88.6% | estimated ± 2.1 pp, high confidence |
| GPT-5.5 | 58.6% | 89.0% | estimated ± 2.1 pp, high confidence |
| GPT-5.6 Luna | 62.7% | 90.8% | estimated ± 2.1 pp, high confidence |
| GPT-5.6 Sol | 64.6% | 91.6% | estimated ± 2.1 pp, high confidence |
| GPT-5.6 Terra | 63.4% | 91.1% | estimated ± 2.1 pp, high confidence |
| Grok 4.5 | 64.7% | 91.6% | estimated ± 2.1 pp, high confidence |
| Hy4 preview | 65.7% | 92.1% | estimated ± 2.1 pp, high confidence |
| Laguna S 2.1 | 59.4% | 89.3% | estimated ± 2.1 pp, high confidence |
| Ling 3.0 Flash | 56.6% | 88.1% | estimated ± 2.1 pp, high confidence |
| MiMo-V2.5 | 56.1% | 87.9% | estimated ± 2.1 pp, high confidence |
| MiMo-V2.5-Pro | 57.2% | 88.3% | estimated ± 2.1 pp, high confidence |
| MiniMax M2.7 | 56.2% | 87.9% | estimated ± 2.1 pp, high confidence |
| Muse Spark 1.1 | 61.5% | 90.2% | estimated ± 2.1 pp, high confidence |
| Qwen 3.6 Max (preview) | 57.3% | 88.4% | estimated ± 2.1 pp, high confidence |
| Qwen3.8 Max | 67.7% | 93.0% | estimated ± 2.1 pp, high confidence |
| Step 3.7 Flash | 56.3% | 87.9% | estimated ± 2.1 pp, high confidence |