Calibration
AA Coding Index → CursorBench 3.2
CursorBench 3.2 is estimated from AA Coding Index with a inverse Michaelis–Menten curve fitted on 17 models measured on both: y = 0.94752·(x − 0.0000) / (0.0000 + 1.8579 − x), R² = 0.72, cross-validated error 4.4 pp. It is used for 53 estimates.
| Estimated model | AA Coding Index | CursorBench 3.2 | Source |
|---|---|---|---|
| Apodex 1.1 | 60.8% | 46.0% | estimated ± 4.4 pp, high confidence |
| Apodex 1.1 Mini | 60.8% | 46.0% | estimated ± 4.4 pp, high confidence |
| Celeris-1 | 14.4% | 8.0% | estimated ± 4.4 pp, medium confidence |
| Claude 3 Opus | 19.5% | 11.1% | estimated ± 4.4 pp, medium confidence |
| Command A+ | 27.9% | 16.7% | estimated ± 4.4 pp, medium confidence |
| DeepSeek V3 | 23.0% | 13.4% | estimated ± 4.4 pp, medium confidence |
| Gemini 1.5 Pro | 23.6% | 13.8% | estimated ± 4.4 pp, medium confidence |
| Gemini 2.5 Pro | 33.3% | 20.7% | estimated ± 4.4 pp, medium confidence |
| Gemini 3.1 Pro | 68.8% | 55.8% | estimated ± 4.4 pp, high confidence |
| Gemma 3 27B | 10.1% | 5.4% | estimated ± 4.4 pp, medium confidence |
| Gemma 4 12B | 31.0% | 18.9% | estimated ± 4.4 pp, medium confidence |
| Gemma 4 26B A4B | 39.3% | 25.4% | estimated ± 4.4 pp, medium confidence |
| Gemma 4 31B | 43.4% | 28.9% | estimated ± 4.4 pp, medium confidence |
| Gemma 4 E2B | 7.2% | 3.8% | estimated ± 4.4 pp, medium confidence |
| Gemma 4 E4B | 9.4% | 5.0% | estimated ± 4.4 pp, medium confidence |
| GLM-4.7 | 45.3% | 30.5% | estimated ± 4.4 pp, medium confidence |
| GPT-4.1 mini | 20.2% | 11.6% | estimated ± 4.4 pp, medium confidence |
| GPT-4.1 nano | 11.1% | 6.0% | estimated ± 4.4 pp, medium confidence |
| GPT-4 Turbo | 21.5% | 12.4% | estimated ± 4.4 pp, medium confidence |
| GPT-4o mini | 11.4% | 6.2% | estimated ± 4.4 pp, medium confidence |
| GPT-5.1 | 49.4% | 34.3% | estimated ± 4.4 pp, medium confidence |
| GPT-5.4 nano | 56.1% | 41.0% | estimated ± 4.4 pp, medium confidence |
| GPT-5 (high) | 37.8% | 24.2% | estimated ± 4.4 pp, medium confidence |
| GPT-OSS 120B | 30.4% | 18.6% | estimated ± 4.4 pp, medium confidence |
| GPT-OSS 20B | 20.7% | 11.9% | estimated ± 4.4 pp, medium confidence |
| Grok 4.3 | 42.3% | 27.9% | estimated ± 4.4 pp, medium confidence |
| Hy3 | 58.8% | 43.9% | estimated ± 4.4 pp, medium confidence |
| Hy3 Preview | 58.8% | 43.9% | estimated ± 4.4 pp, medium confidence |
| K-Exaone | 32.1% | 19.8% | estimated ± 4.4 pp, medium confidence |
| Kimi K2.5 (Reasoning) | 46.8% | 31.9% | estimated ± 4.4 pp, medium confidence |
| LFM2.5-2.6B | 7.7% | 4.1% | estimated ± 4.4 pp, medium confidence |
| Ling 2.6 Flash | 25.3% | 14.9% | estimated ± 4.4 pp, medium confidence |
| Ling 3.0 Flash FP8 | 50.6% | 35.5% | estimated ± 4.4 pp, medium confidence |
| Llama 4 Maverick | 16.3% | 9.1% | estimated ± 4.4 pp, medium confidence |
| Llama 4 Scout | 8.2% | 4.4% | estimated ± 4.4 pp, medium confidence |
| MiMo-V2-Flash | 49.8% | 34.7% | estimated ± 4.4 pp, medium confidence |
| Mistral Large 3 | 20.1% | 11.5% | estimated ± 4.4 pp, medium confidence |
| Mistral Medium 3.5 128B | 46.9% | 32.0% | estimated ± 4.4 pp, medium confidence |
| Mistral Small 4 | 26.6% | 15.9% | estimated ± 4.4 pp, medium confidence |
| Mistral Small 4 (Reasoning) | 26.6% | 15.9% | estimated ± 4.4 pp, medium confidence |
| Nemotron 3.5 Lightning 30B A3B NVFP4 | 26.8% | 15.9% | estimated ± 4.4 pp, medium confidence |
| Nemotron 3 Nano 30B | 14.4% | 7.9% | estimated ± 4.4 pp, medium confidence |
| Nemotron 3 Nano Omni 30B A3B | 13.8% | 7.6% | estimated ± 4.4 pp, medium confidence |
| Nemotron 3 Super 100B | 37.7% | 24.1% | estimated ± 4.4 pp, medium confidence |
| Nemotron 3 Ultra | 49.3% | 34.2% | estimated ± 4.4 pp, medium confidence |
| o1 | 39.7% | 25.8% | estimated ± 4.4 pp, medium confidence |
| o1-preview | 34.1% | 21.3% | estimated ± 4.4 pp, medium confidence |
| Quasar 438B | 61.2% | 46.5% | estimated ± 4.4 pp, high confidence |
| Qwen3.5-122B-A10B | 45.7% | 30.9% | estimated ± 4.4 pp, medium confidence |
| Qwen3.8 Max Preview | 71.8% | 59.7% | estimated ± 4.4 pp, high confidence |
| Trinity-Large-Preview | 25.8% | 15.3% | estimated ± 4.4 pp, medium confidence |
| Trinity-Large-Thinking | 25.8% | 15.3% | estimated ± 4.4 pp, medium confidence |
| Ultravox v0.6 Llama 3.3 70B | 11.9% | 6.5% | estimated ± 4.4 pp, medium confidence |