Calibration
Vals GPQA Diamond → ARC-AGI-2
ARC-AGI-2 is estimated from Vals GPQA Diamond with a linear curve fitted on 24 models measured on both: y = 4.3298·x + -3.2871, R² = 0.72, cross-validated error 12.6 pp. It is used for 14 estimates.
| Estimated model | Vals GPQA Diamond | ARC-AGI-2 | Source |
|---|---|---|---|
| MiniMax M3 | 92.7% | 72.7% | estimated ± 12.6 pp, low confidence |
| Muse Spark 1.1 | 91.2% | 66.2% | estimated ± 12.6 pp, low confidence |
| Claude Sonnet 5 | 88.9% | 56.2% | estimated ± 12.6 pp, low confidence |
| GLM-5.3 | 88.1% | 52.7% | estimated ± 12.6 pp, low confidence |
| Gemini 3 Flash | 87.9% | 51.9% | estimated ± 12.6 pp, low confidence |
| GLM-5.3-Flash | 86.4% | 45.4% | estimated ± 12.6 pp, low confidence |
| Gemini 3.5 Flash-Lite | 83.8% | 34.1% | estimated ± 12.6 pp, low confidence |
| Qwen3.5 Flash | 82.8% | 29.8% | estimated ± 12.6 pp, low confidence |
| MiMo-V2.5-Pro | 82.6% | 28.9% | estimated ± 12.6 pp, low confidence |
| MiMo-V2.5 | 81.6% | 24.6% | estimated ± 12.6 pp, low confidence |
| Gemini 3.1 Flash-Lite | 81.1% | 22.4% | estimated ± 12.6 pp, low confidence |
| Claude Haiku 4.5 | 72.2% | 0.0% | estimated ± 12.6 pp, low confidence |
| GLM-4.5 | 72.2% | 0.0% | estimated ± 12.6 pp, low confidence |
| GLM-4.6 | 74.5% | 0.0% | estimated ± 12.6 pp, low confidence |
Fitted on
| Model | Vals GPQA Diamond | ARC-AGI-2 |
|---|---|---|
| Claude Fable 5 | 93.2% | 89.2% |
| Claude Fable 5.1 | 93.4% | 90.0% |
| Claude Opus 4.8 | 92.4% | 72.1% |
| Claude Opus 5 | 93.4% | 90.4% |
| Claude Sonnet 4.6 | 85.6% | 58.3% |
| DeepSeek V4 Flash 0731 | 89.9% | 61.4% |
| DeepSeek V4 Pro 0813 | 92.4% | 61.3% |
| Gemini 3.1 Pro | 95.5% | 77.1% |
| Gemini 3.5 Flash | 92.7% | 72.1% |
| Gemini 3.6 Flash | 93.4% | 60.4% |
| Gemini 3.7 Flash | 93.9% | 84.6% |
| Gemini 3.8 Flash | 94.4% | 89.2% |
| GPT-5.4 mini | 83.1% | 18.9% |
| GPT-5.4 nano | 77.5% | 5.7% |
| GPT-5.5 | 93.2% | 85.0% |
| GPT-5.6 Luna | 91.7% | 59.5% |
| GPT-5.6 Sol | 95.2% | 92.5% |
| GPT-5.6 Terra | 90.9% | 83.9% |
| Grok 4.20 | 88.6% | 53.3% |
| Grok 4.5 | 92.9% | 52.6% |
| Grok 4.6 | 94.7% | 67.1% |
| Inkling-Small | 83.6% | 40.1% |
| Kimi K3 | 92.9% | 60.4% |
| Muse Spark | 89.6% | 42.5% |