Calibration
GDPval-AA → JobBench
JobBench is estimated from GDPval-AA with a linear curve fitted on 13 models measured on both: y = 1.2398·x + -0.1001, R² = 0.83, cross-validated error 8.6 pp. It is used for 61 estimates.
| Estimated model | GDPval-AA | JobBench | Source |
|---|---|---|---|
| A.X K2 | 23.2% | 18.8% | estimated ± 8.6 pp, medium confidence |
| Celeris-1 | 0.0% | 0.0% | estimated ± 8.6 pp, low confidence |
| Claude Fable 5 | 55.6% | 58.9% | estimated ± 8.6 pp, medium confidence |
| Claude Sonnet 5 | 48.3% | 49.9% | estimated ± 8.6 pp, medium confidence |
| Command A+ | 0.9% | 0.0% | estimated ± 8.6 pp, low confidence |
| DeepSeek V3 | 0.0% | 0.0% | estimated ± 8.6 pp, low confidence |
| DeepSeek V3 0324 | 0.0% | 0.0% | estimated ± 8.6 pp, low confidence |
| DeepSeek V4.1 Flash | 55.0% | 58.2% | estimated ± 8.6 pp, medium confidence |
| Gemini 2.5 Pro | 0.0% | 0.0% | estimated ± 8.6 pp, low confidence |
| Gemini 3.5 Flash-Lite | 24.3% | 20.1% | estimated ± 8.6 pp, medium confidence |
| Gemini 3.6 Flash | 39.3% | 38.7% | estimated ± 8.6 pp, medium confidence |
| Gemma 3 27B | 0.0% | 0.0% | estimated ± 8.6 pp, low confidence |
| Gemma 4 12B | 0.0% | 0.0% | estimated ± 8.6 pp, low confidence |
| Gemma 4 26B A4B | 3.4% | 0.0% | estimated ± 8.6 pp, low confidence |
| Gemma 4 31B | 6.1% | 0.0% | estimated ± 8.6 pp, low confidence |
| Gemma 4 E2B | 0.0% | 0.0% | estimated ± 8.6 pp, low confidence |
| Gemma 4 E4B | 0.0% | 0.0% | estimated ± 8.6 pp, low confidence |
| GLM-4.7 | 25.8% | 22.0% | estimated ± 8.6 pp, medium confidence |
| GPT-4.1 mini | 0.0% | 0.0% | estimated ± 8.6 pp, low confidence |
| GPT-4.1 nano | 0.0% | 0.0% | estimated ± 8.6 pp, low confidence |
| GPT-4o | 0.0% | 0.0% | estimated ± 8.6 pp, low confidence |
| GPT-4o mini | 0.0% | 0.0% | estimated ± 8.6 pp, low confidence |
| GPT-5.1 | 16.5% | 10.4% | estimated ± 8.6 pp, low confidence |
| GPT-6.1 Sol | 53.8% | 56.7% | estimated ± 8.6 pp, medium confidence |
| GPT-6 Luna | 46.9% | 48.1% | estimated ± 8.6 pp, medium confidence |
| Granite 4.2 30B | 3.9% | 0.0% | estimated ± 8.6 pp, low confidence |
| Granite 4.2 3B | 0.0% | 0.0% | estimated ± 8.6 pp, low confidence |
| Granite 4.2 8B | 0.0% | 0.0% | estimated ± 8.6 pp, low confidence |
| Grok 4.5 | 44.5% | 45.2% | estimated ± 8.6 pp, medium confidence |
| Grok 4.6 | 56.1% | 59.5% | estimated ± 8.6 pp, medium confidence |
| Grok 4.7 | 60.8% | 65.4% | estimated ± 8.6 pp, low confidence |
| Hy3 | 28.3% | 25.1% | estimated ± 8.6 pp, medium confidence |
| Hy3 Preview | 35.8% | 34.4% | estimated ± 8.6 pp, medium confidence |
| K-Exaone | 0.0% | 0.0% | estimated ± 8.6 pp, low confidence |
| Ling 2.6 Flash | 0.0% | 0.0% | estimated ± 8.6 pp, low confidence |
| Ling 3.0 Flash FP8 | 22.4% | 17.8% | estimated ± 8.6 pp, medium confidence |
| Ling 3.0 Flash VL | 33.2% | 31.1% | estimated ± 8.6 pp, medium confidence |
| Ling 3.0 Tiny | 3.2% | 0.0% | estimated ± 8.6 pp, low confidence |
| Ling 3.1 Flash | 56.1% | 59.5% | estimated ± 8.6 pp, medium confidence |
| Llama 4 Maverick | 0.0% | 0.0% | estimated ± 8.6 pp, low confidence |
| Llama 4 Scout | 0.0% | 0.0% | estimated ± 8.6 pp, low confidence |
| Mercury 2.5 | 0.0% | 0.0% | estimated ± 8.6 pp, low confidence |
| MiMo-V2-Flash | 6.6% | 0.0% | estimated ± 8.6 pp, low confidence |
| MiniCPM5-2B | 10.5% | 3.0% | estimated ± 8.6 pp, low confidence |
| Mistral Large 3 | 0.0% | 0.0% | estimated ± 8.6 pp, low confidence |
| Mistral Large 4 | 46.2% | 47.3% | estimated ± 8.6 pp, medium confidence |
| Mistral Medium 3.5 128B | 13.2% | 6.4% | estimated ± 8.6 pp, low confidence |
| Mistral Small 4 | 0.0% | 0.0% | estimated ± 8.6 pp, low confidence |
| Mistral Small 4 (Reasoning) | 0.0% | 0.0% | estimated ± 8.6 pp, low confidence |
| Muse Spark 1.2 | 48.9% | 50.6% | estimated ± 8.6 pp, medium confidence |
| Nemotron 3.5 Lightning 30B A3B NVFP4 | 7.0% | 0.0% | estimated ± 8.6 pp, low confidence |
| Nemotron 3 Nano 30B | 0.0% | 0.0% | estimated ± 8.6 pp, low confidence |
| Nemotron 3 Nano Omni 30B A3B | 0.0% | 0.0% | estimated ± 8.6 pp, low confidence |
| Nemotron 3 Ultra | 33.1% | 31.0% | estimated ± 8.6 pp, medium confidence |
| North Mini Code | 0.0% | 0.0% | estimated ± 8.6 pp, low confidence |
| Quasar 438B | 34.2% | 32.4% | estimated ± 8.6 pp, medium confidence |
| Qwen3.5-122B-A10B | 15.8% | 9.6% | estimated ± 8.6 pp, low confidence |
| Solar Pro 3 | 0.0% | 0.0% | estimated ± 8.6 pp, low confidence |
| Trinity-Large-Preview | 0.0% | 0.0% | estimated ± 8.6 pp, low confidence |
| Trinity-Large-Thinking | 0.0% | 0.0% | estimated ± 8.6 pp, low confidence |
| Ultravox v0.6 Llama 3.3 70B | 0.0% | 0.0% | estimated ± 8.6 pp, low confidence |