Calibration
AA Coding Index → LiveCodeBench
LiveCodeBench is estimated from AA Coding Index with a inverse Michaelis–Menten curve fitted on 6 models measured on both: y = 3.02770·(x − 0.0077) / (0.0077 + 2.0000 − x), R² = 0.78, cross-validated error 8.9 pp. It is used for 96 estimates.
| Estimated model | AA Coding Index | LiveCodeBench | Source |
|---|---|---|---|
| Apodex 1.1 | 60.8% | 100.0% | estimated ± 8.9 pp, low confidence |
| Apodex 1.1 Mini | 60.8% | 100.0% | estimated ± 8.9 pp, low confidence |
| Celeris-1 | 14.4% | 22.1% | estimated ± 8.9 pp, low confidence |
| Claude 3 Opus | 19.5% | 31.3% | estimated ± 8.9 pp, low confidence |
| Claude Fable 5 | 76.5% | 100.0% | estimated ± 8.9 pp, low confidence |
| Claude Fable 5.1 | 81.6% | 100.0% | estimated ± 8.9 pp, low confidence |
| Claude Opus 4.7 (Adaptive) | 73.6% | 100.0% | estimated ± 8.9 pp, low confidence |
| Claude Opus 4.8 | 74.3% | 100.0% | estimated ± 8.9 pp, low confidence |
| Claude Opus 5 | 78.0% | 100.0% | estimated ± 8.9 pp, low confidence |
| Claude Sonnet 5 | 71.6% | 100.0% | estimated ± 8.9 pp, low confidence |
| Command A+ | 27.9% | 47.4% | estimated ± 8.9 pp, low confidence |
| DeepSeek V4 Flash 0731 | 69.1% | 100.0% | estimated ± 8.9 pp, low confidence |
| DeepSeek V4 Pro 0813 | 68.8% | 100.0% | estimated ± 8.9 pp, low confidence |
| Gemini 1.5 Pro | 23.6% | 39.1% | estimated ± 8.9 pp, low confidence |
| Gemini 2.5 Pro | 33.3% | 58.7% | estimated ± 8.9 pp, low confidence |
| Gemini 3.1 Pro | 68.8% | 100.0% | estimated ± 8.9 pp, low confidence |
| Gemini 3.5 Flash | 70.1% | 100.0% | estimated ± 8.9 pp, low confidence |
| Gemini 3.5 Flash-Lite | 49.3% | 97.1% | estimated ± 8.9 pp, low confidence |
| Gemini 3.6 Flash | 69.2% | 100.0% | estimated ± 8.9 pp, low confidence |
| Gemini 3.7 Flash | 76.1% | 100.0% | estimated ± 8.9 pp, low confidence |
| Gemini 3.8 Flash | 76.3% | 100.0% | estimated ± 8.9 pp, low confidence |
| Gemma 3 27B | 10.1% | 14.7% | estimated ± 8.9 pp, low confidence |
| Gemma 4 12B | 31.0% | 53.8% | estimated ± 8.9 pp, low confidence |
| Gemma 4 26B A4B | 39.3% | 72.3% | estimated ± 8.9 pp, low confidence |
| Gemma 4 31B | 43.4% | 82.1% | estimated ± 8.9 pp, low confidence |
| Gemma 4 E2B | 7.2% | 10.1% | estimated ± 8.9 pp, low confidence |
| Gemma 4 E4B | 9.4% | 13.6% | estimated ± 8.9 pp, low confidence |
| GLM-5.1 | 55.8% | 100.0% | estimated ± 8.9 pp, low confidence |
| GLM-5.2 | 68.8% | 100.0% | estimated ± 8.9 pp, low confidence |
| GLM-5.3 | 74.8% | 100.0% | estimated ± 8.9 pp, low confidence |
| GPT-4.1 mini | 20.2% | 32.6% | estimated ± 8.9 pp, low confidence |
| GPT-4.1 nano | 11.1% | 16.6% | estimated ± 8.9 pp, low confidence |
| GPT-4 Turbo | 21.5% | 35.0% | estimated ± 8.9 pp, low confidence |
| GPT-4o mini | 11.4% | 17.0% | estimated ± 8.9 pp, low confidence |
| GPT-5.1 | 49.4% | 97.2% | estimated ± 8.9 pp, low confidence |
| GPT-5.4 | 71.1% | 100.0% | estimated ± 8.9 pp, low confidence |
| GPT-5.4 mini | 56.1% | 100.0% | estimated ± 8.9 pp, low confidence |
| GPT-5.4 nano | 56.1% | 100.0% | estimated ± 8.9 pp, low confidence |
| GPT-5.5 | 74.9% | 100.0% | estimated ± 8.9 pp, low confidence |
| GPT-5.6 Luna | 71.5% | 100.0% | estimated ± 8.9 pp, low confidence |
| GPT-5.6 Sol | 77.4% | 100.0% | estimated ± 8.9 pp, low confidence |
| GPT-5.6 Terra | 76.7% | 100.0% | estimated ± 8.9 pp, low confidence |
| GPT-5 (high) | 37.8% | 68.7% | estimated ± 8.9 pp, low confidence |
| GPT-6 Astra | 76.9% | 100.0% | estimated ± 8.9 pp, low confidence |
| GPT-OSS 120B | 30.4% | 52.7% | estimated ± 8.9 pp, low confidence |
| GPT-OSS 20B | 20.7% | 33.5% | estimated ± 8.9 pp, low confidence |
| Granite 4.2 8B | 22.4% | 36.7% | estimated ± 8.9 pp, low confidence |
| Grok 4.3 | 42.3% | 79.2% | estimated ± 8.9 pp, low confidence |
| Grok 4.5 | 72.5% | 100.0% | estimated ± 8.9 pp, low confidence |
| Grok 4.6 | 76.8% | 100.0% | estimated ± 8.9 pp, low confidence |
| Hy3 | 58.8% | 100.0% | estimated ± 8.9 pp, low confidence |
| Hy3 Preview | 58.8% | 100.0% | estimated ± 8.9 pp, low confidence |
| Inkling | 52.1% | 100.0% | estimated ± 8.9 pp, low confidence |
| Inkling-Small | 52.9% | 100.0% | estimated ± 8.9 pp, low confidence |
| K-Exaone | 32.1% | 56.3% | estimated ± 8.9 pp, low confidence |
| Kimi K2.6 | 61.8% | 100.0% | estimated ± 8.9 pp, low confidence |
| Kimi K2.5 | 46.8% | 90.5% | estimated ± 8.9 pp, low confidence |
| Kimi K2.5 (Reasoning) | 46.8% | 90.5% | estimated ± 8.9 pp, low confidence |
| Kimi K2.7 Code | 60.8% | 100.0% | estimated ± 8.9 pp, low confidence |
| Kimi K3 | 76.2% | 100.0% | estimated ± 8.9 pp, low confidence |
| LFM2.5-2.6B | 7.7% | 10.9% | estimated ± 8.9 pp, low confidence |
| Ling 2.6 Flash | 25.3% | 42.2% | estimated ± 8.9 pp, low confidence |
| Ling 3.0 Flash | 50.6% | 100.0% | estimated ± 8.9 pp, low confidence |
| Ling 3.0 Flash FP8 | 50.6% | 100.0% | estimated ± 8.9 pp, low confidence |
| Llama 4 Maverick | 16.3% | 25.5% | estimated ± 8.9 pp, low confidence |
| Llama 4 Scout | 8.2% | 11.6% | estimated ± 8.9 pp, low confidence |
| MiMo-V2.5-Pro | 60.2% | 100.0% | estimated ± 8.9 pp, low confidence |
| MiMo-V2-Flash | 49.8% | 98.4% | estimated ± 8.9 pp, low confidence |
| MiniMax M2.7 | 52.6% | 100.0% | estimated ± 8.9 pp, low confidence |
| MiniMax M3 | 58.6% | 100.0% | estimated ± 8.9 pp, low confidence |
| Mistral Large 3 | 20.1% | 32.3% | estimated ± 8.9 pp, low confidence |
| Mistral Medium 3.5 128B | 46.9% | 90.8% | estimated ± 8.9 pp, low confidence |
| Mistral Small 4 | 26.6% | 45.0% | estimated ± 8.9 pp, low confidence |
| Mistral Small 4 (Reasoning) | 26.6% | 45.0% | estimated ± 8.9 pp, low confidence |
| Muse Glimmer 30B | 49.0% | 96.2% | estimated ± 8.9 pp, low confidence |
| Muse Spark | 58.6% | 100.0% | estimated ± 8.9 pp, low confidence |
| Muse Spark 1.1 | 71.3% | 100.0% | estimated ± 8.9 pp, low confidence |
| Muse Spark 1.2 | 72.2% | 100.0% | estimated ± 8.9 pp, low confidence |
| Muse Spark 1.3 | 75.8% | 100.0% | estimated ± 8.9 pp, low confidence |
| Nemotron 3.5 Lightning 30B A3B NVFP4 | 26.8% | 45.2% | estimated ± 8.9 pp, low confidence |
| Nemotron 3 Nano 30B | 14.4% | 22.1% | estimated ± 8.9 pp, low confidence |
| Nemotron 3 Nano Omni 30B A3B | 13.8% | 21.0% | estimated ± 8.9 pp, low confidence |
| Nemotron 3 Super 100B | 37.7% | 68.6% | estimated ± 8.9 pp, low confidence |
| Nemotron 3 Ultra | 49.3% | 96.9% | estimated ± 8.9 pp, low confidence |
| o1 | 39.7% | 73.2% | estimated ± 8.9 pp, low confidence |
| o1-preview | 34.1% | 60.4% | estimated ± 8.9 pp, low confidence |
| Quasar 438B | 61.2% | 100.0% | estimated ± 8.9 pp, low confidence |
| Qwen3.5-122B-A10B | 45.7% | 87.7% | estimated ± 8.9 pp, low confidence |
| Qwen3.6 Plus | 54.5% | 100.0% | estimated ± 8.9 pp, low confidence |
| Qwen3.8-27B | 68.1% | 100.0% | estimated ± 8.9 pp, low confidence |
| Qwen3.8-Flash-Next | 73.1% | 100.0% | estimated ± 8.9 pp, low confidence |
| Qwen3.8 Max Preview | 71.8% | 100.0% | estimated ± 8.9 pp, low confidence |
| Step 3.7 Flash | 39.6% | 72.9% | estimated ± 8.9 pp, low confidence |
| Trinity-Large-Preview | 25.8% | 43.3% | estimated ± 8.9 pp, low confidence |
| Trinity-Large-Thinking | 25.8% | 43.3% | estimated ± 8.9 pp, low confidence |
| Ultravox v0.6 Llama 3.3 70B | 11.9% | 17.9% | estimated ± 8.9 pp, low confidence |