Calibration
AA Coding Index → DeepSWE
DeepSWE is estimated from AA Coding Index with a Hill curve fitted on 20 models measured on both: y = 0.0000 + (1.0062 − 0.0000)·x^6.00 / (0.67629^6.00 + x^6.00), R² = 0.61, cross-validated error 6.4 pp. It is used for 73 estimates.
| Estimated model | AA Coding Index | DeepSWE | Source |
|---|---|---|---|
| Apodex 1.1 | 60.8% | 34.7% | estimated ± 6.4 pp, low confidence |
| Apodex 1.1 Mini | 60.8% | 34.7% | estimated ± 6.4 pp, low confidence |
| Celeris-1 | 14.4% | 0.0% | estimated ± 6.4 pp, low confidence |
| Claude 3 Opus | 19.5% | 0.1% | estimated ± 6.4 pp, low confidence |
| Claude Opus 4.7 (Adaptive) | 73.6% | 62.8% | estimated ± 6.4 pp, medium confidence |
| Command A+ | 27.9% | 0.5% | estimated ± 6.4 pp, low confidence |
| DeepSeek V3 | 23.0% | 0.2% | estimated ± 6.4 pp, low confidence |
| Gemini 1.5 Pro | 23.6% | 0.2% | estimated ± 6.4 pp, low confidence |
| Gemini 2.5 Pro | 33.3% | 1.4% | estimated ± 6.4 pp, low confidence |
| Gemini 3.5 Flash | 70.1% | 55.8% | estimated ± 6.4 pp, medium confidence |
| Gemini 3.5 Flash-Lite | 49.3% | 13.2% | estimated ± 6.4 pp, low confidence |
| Gemma 3 27B | 10.1% | 0.0% | estimated ± 6.4 pp, low confidence |
| Gemma 4 12B | 31.0% | 0.9% | estimated ± 6.4 pp, low confidence |
| Gemma 4 26B A4B | 39.3% | 3.7% | estimated ± 6.4 pp, low confidence |
| Gemma 4 31B | 43.4% | 6.6% | estimated ± 6.4 pp, low confidence |
| Gemma 4 E2B | 7.2% | 0.0% | estimated ± 6.4 pp, low confidence |
| Gemma 4 E4B | 9.4% | 0.0% | estimated ± 6.4 pp, low confidence |
| GLM-4.7 | 45.3% | 8.3% | estimated ± 6.4 pp, low confidence |
| GLM-5.1 | 55.8% | 24.1% | estimated ± 6.4 pp, low confidence |
| GPT-4.1 mini | 20.2% | 0.1% | estimated ± 6.4 pp, low confidence |
| GPT-4.1 nano | 11.1% | 0.0% | estimated ± 6.4 pp, low confidence |
| GPT-4 Turbo | 21.5% | 0.1% | estimated ± 6.4 pp, low confidence |
| GPT-4o mini | 11.4% | 0.0% | estimated ± 6.4 pp, low confidence |
| GPT-5.1 | 49.4% | 13.3% | estimated ± 6.4 pp, low confidence |
| GPT-5.4 nano | 56.1% | 24.7% | estimated ± 6.4 pp, low confidence |
| GPT-5 (high) | 37.8% | 3.0% | estimated ± 6.4 pp, low confidence |
| GPT-OSS 120B | 30.4% | 0.8% | estimated ± 6.4 pp, low confidence |
| GPT-OSS 20B | 20.7% | 0.1% | estimated ± 6.4 pp, low confidence |
| Granite 4.2 8B | 22.4% | 0.1% | estimated ± 6.4 pp, low confidence |
| Grok 4.3 | 42.3% | 5.6% | estimated ± 6.4 pp, low confidence |
| Hy3 | 58.8% | 30.4% | estimated ± 6.4 pp, low confidence |
| Hy3 Preview | 58.8% | 30.4% | estimated ± 6.4 pp, low confidence |
| Inkling-Small | 52.9% | 18.8% | estimated ± 6.4 pp, low confidence |
| K-Exaone | 32.1% | 1.1% | estimated ± 6.4 pp, low confidence |
| Kimi K2.6 | 61.8% | 37.0% | estimated ± 6.4 pp, low confidence |
| Kimi K2.5 | 46.8% | 9.9% | estimated ± 6.4 pp, low confidence |
| Kimi K2.5 (Reasoning) | 46.8% | 9.9% | estimated ± 6.4 pp, low confidence |
| Kimi K2.7 Code | 60.8% | 34.7% | estimated ± 6.4 pp, low confidence |
| LFM2.5-2.6B | 7.7% | 0.0% | estimated ± 6.4 pp, low confidence |
| Ling 2.6 Flash | 25.3% | 0.3% | estimated ± 6.4 pp, low confidence |
| Ling 3.0 Flash | 50.6% | 15.1% | estimated ± 6.4 pp, low confidence |
| Ling 3.0 Flash FP8 | 50.6% | 15.1% | estimated ± 6.4 pp, low confidence |
| Llama 4 Maverick | 16.3% | 0.0% | estimated ± 6.4 pp, low confidence |
| Llama 4 Scout | 8.2% | 0.0% | estimated ± 6.4 pp, low confidence |
| MiMo-V2.5-Pro | 60.2% | 33.4% | estimated ± 6.4 pp, low confidence |
| MiMo-V2-Flash | 49.8% | 13.9% | estimated ± 6.4 pp, low confidence |
| MiniMax M2.7 | 52.6% | 18.3% | estimated ± 6.4 pp, low confidence |
| MiniMax M3 | 58.6% | 29.9% | estimated ± 6.4 pp, low confidence |
| Mistral Large 3 | 20.1% | 0.1% | estimated ± 6.4 pp, low confidence |
| Mistral Medium 3.5 128B | 46.9% | 10.1% | estimated ± 6.4 pp, low confidence |
| Mistral Small 4 | 26.6% | 0.4% | estimated ± 6.4 pp, low confidence |
| Mistral Small 4 (Reasoning) | 26.6% | 0.4% | estimated ± 6.4 pp, low confidence |
| Muse Glimmer 30B | 49.0% | 12.7% | estimated ± 6.4 pp, low confidence |
| Muse Spark | 58.6% | 30.0% | estimated ± 6.4 pp, low confidence |
| Nemotron 3.5 Lightning 30B A3B NVFP4 | 26.8% | 0.4% | estimated ± 6.4 pp, low confidence |
| Nemotron 3 Nano 30B | 14.4% | 0.0% | estimated ± 6.4 pp, low confidence |
| Nemotron 3 Nano Omni 30B A3B | 13.8% | 0.0% | estimated ± 6.4 pp, low confidence |
| Nemotron 3 Super 100B | 37.7% | 2.9% | estimated ± 6.4 pp, low confidence |
| Nemotron 3 Ultra | 49.3% | 13.1% | estimated ± 6.4 pp, low confidence |
| o1 | 39.7% | 4.0% | estimated ± 6.4 pp, low confidence |
| o1-preview | 34.1% | 1.6% | estimated ± 6.4 pp, low confidence |
| Quasar 438B | 61.2% | 35.7% | estimated ± 6.4 pp, low confidence |
| Qwen3.5-122B-A10B | 45.7% | 8.8% | estimated ± 6.4 pp, low confidence |
| Qwen3.6-27B | 53.7% | 20.2% | estimated ± 6.4 pp, low confidence |
| Qwen3.6-35B-A3B | 41.9% | 5.4% | estimated ± 6.4 pp, low confidence |
| Qwen3.6 Plus | 54.5% | 21.7% | estimated ± 6.4 pp, low confidence |
| Qwen3.7 Max | 66.0% | 46.6% | estimated ± 6.4 pp, low confidence |
| Qwen3.7 Plus | 55.9% | 24.3% | estimated ± 6.4 pp, low confidence |
| Qwen3.8 Max Preview | 71.8% | 59.3% | estimated ± 6.4 pp, medium confidence |
| Step 3.7 Flash | 39.6% | 3.9% | estimated ± 6.4 pp, low confidence |
| Trinity-Large-Preview | 25.8% | 0.3% | estimated ± 6.4 pp, low confidence |
| Trinity-Large-Thinking | 25.8% | 0.3% | estimated ± 6.4 pp, low confidence |
| Ultravox v0.6 Llama 3.3 70B | 11.9% | 0.0% | estimated ± 6.4 pp, low confidence |