Calibration
CritPt → ARC-AGI-1
ARC-AGI-1 is estimated from CritPt with a linear curve fitted on 37 models measured on both: y = 0.7491·x + 0.7474, R² = 0.46, cross-validated error 7.8 pp. It is used for 88 estimates.
| Estimated model | CritPt | ARC-AGI-1 | Source |
|---|---|---|---|
| Muse Wonder Full | 32.9% | 99.4% | estimated ± 7.8 pp, low confidence |
| Muse Wonder | 28.3% | 95.9% | estimated ± 7.8 pp, low confidence |
| MiMo-V2.6-Pro | 26.6% | 94.7% | estimated ± 7.8 pp, low confidence |
| Muse Spark 1.3 | 24.9% | 93.4% | estimated ± 7.8 pp, low confidence |
| Sage-A1.5 | 20.0% | 89.7% | estimated ± 7.8 pp, low confidence |
| Ling 3.1 Flash | 18.0% | 88.2% | estimated ± 7.8 pp, low confidence |
| Muse Spark 1.2 | 17.7% | 88.0% | estimated ± 7.8 pp, low confidence |
| Qwen3.8 Max Preview | 17.7% | 88.0% | estimated ± 7.8 pp, low confidence |
| GPT-5.3 Codex | 16.9% | 87.4% | estimated ± 7.8 pp, low confidence |
| Muse Spark Open | 14.9% | 85.9% | estimated ± 7.8 pp, low confidence |
| MiMo-V2.6-Flash | 12.0% | 83.7% | estimated ± 7.8 pp, low confidence |
| Kimi K2.7 Code | 10.0% | 82.2% | estimated ± 7.8 pp, low confidence |
| Mistral Large 4 | 10.0% | 82.2% | estimated ± 7.8 pp, low confidence |
| Quasar 438B | 9.4% | 81.8% | estimated ± 7.8 pp, low confidence |
| GPT-5.2-Codex | 8.7% | 81.3% | estimated ± 7.8 pp, low confidence |
| GPT-5.1-Codex | 5.7% | 79.0% | estimated ± 7.8 pp, low confidence |
| GPT-5.1-Codex-Max | 5.7% | 79.0% | estimated ± 7.8 pp, low confidence |
| Hy3 | 4.9% | 78.4% | estimated ± 7.8 pp, low confidence |
| Apodex 1.1 | 4.6% | 78.2% | estimated ± 7.8 pp, low confidence |
| Apodex 1.1 Mini | 4.6% | 78.2% | estimated ± 7.8 pp, low confidence |
| Qwen 3.6 Max (preview) | 3.7% | 77.5% | estimated ± 7.8 pp, low confidence |
| Muse Glimmer 30B | 2.6% | 76.7% | estimated ± 7.8 pp, low confidence |
| Step 3.7 Flash | 2.3% | 76.5% | estimated ± 7.8 pp, low confidence |
| Ling 3.0 Flash VL | 2.0% | 76.2% | estimated ± 7.8 pp, low confidence |
| Gemini 2.5 Flash | 1.4% | 75.8% | estimated ± 7.8 pp, low confidence |
| Claude 4 Sonnet | 1.1% | 75.6% | estimated ± 7.8 pp, low confidence |
| K-Exaone | 1.1% | 75.6% | estimated ± 7.8 pp, low confidence |
| MiMo-V2-Omni | 1.1% | 75.6% | estimated ± 7.8 pp, low confidence |
| DeepSeek V3.2 | 0.9% | 75.4% | estimated ± 7.8 pp, low confidence |
| Trinity-Large-Preview | 0.9% | 75.4% | estimated ± 7.8 pp, low confidence |
| Trinity-Large-Thinking | 0.9% | 75.4% | estimated ± 7.8 pp, low confidence |
| Gemma 4 E4B | 0.6% | 75.2% | estimated ± 7.8 pp, low confidence |
| GLM-5V-Turbo | 0.6% | 75.2% | estimated ± 7.8 pp, low confidence |
| Command A+ | 0.3% | 75.0% | estimated ± 7.8 pp, low confidence |
| GLM-5-Turbo | 0.3% | 75.0% | estimated ± 7.8 pp, low confidence |
| Granite 4.2 30B | 0.3% | 75.0% | estimated ± 7.8 pp, low confidence |
| Granite 4.2 8B | 0.3% | 75.0% | estimated ± 7.8 pp, low confidence |
| MiMo-V2-Pro | 0.3% | 75.0% | estimated ± 7.8 pp, low confidence |
| MiniCPM5-2B | 0.3% | 75.0% | estimated ± 7.8 pp, low confidence |
| Mistral Small 4 | 0.3% | 75.0% | estimated ± 7.8 pp, low confidence |
| North Mini Code | 0.3% | 75.0% | estimated ± 7.8 pp, low confidence |
| Sarvam 30B | 0.3% | 75.0% | estimated ± 7.8 pp, low confidence |
| Celeris-1 | 0.0% | 74.7% | estimated ± 7.8 pp, low confidence |
| Claude 3.5 Haiku | 0.0% | 74.7% | estimated ± 7.8 pp, low confidence |
| Claude 3 Haiku | 0.0% | 74.7% | estimated ± 7.8 pp, low confidence |
| DeepSeek V3 | 0.0% | 74.7% | estimated ± 7.8 pp, low confidence |
| DeepSeek V3 0324 | 0.0% | 74.7% | estimated ± 7.8 pp, low confidence |
| Devstral Small | 0.0% | 74.7% | estimated ± 7.8 pp, low confidence |
| Exaone 4.0 1.2B | 0.0% | 74.7% | estimated ± 7.8 pp, low confidence |
| Exaone 4.0 32B | 0.0% | 74.7% | estimated ± 7.8 pp, low confidence |
| Gemini 2.5 Flash-Lite | 0.0% | 74.7% | estimated ± 7.8 pp, low confidence |
| Gemma 3 27B | 0.0% | 74.7% | estimated ± 7.8 pp, low confidence |
| Gemma 4 E2B | 0.0% | 74.7% | estimated ± 7.8 pp, low confidence |
| GLM-4.5-Air | 0.0% | 74.7% | estimated ± 7.8 pp, low confidence |
| GPT-4.1 nano | 0.0% | 74.7% | estimated ± 7.8 pp, low confidence |
| GPT-4o | 0.0% | 74.7% | estimated ± 7.8 pp, low confidence |
| Granite-4.0-350M | 0.0% | 74.7% | estimated ± 7.8 pp, low confidence |
| Granite-4.0-H-1B | 0.0% | 74.7% | estimated ± 7.8 pp, low confidence |
| Granite-4.0-H-350M | 0.0% | 74.7% | estimated ± 7.8 pp, low confidence |
| Granite 4.2 3B | 0.0% | 74.7% | estimated ± 7.8 pp, low confidence |
| Grok 4.1 Fast | 0.0% | 74.7% | estimated ± 7.8 pp, low confidence |
| Grok Code Fast 1 | 0.0% | 74.7% | estimated ± 7.8 pp, low confidence |
| LFM2.5-2.6B | 0.0% | 74.7% | estimated ± 7.8 pp, low confidence |
| LFM2.5-8B-A1B | 0.0% | 74.7% | estimated ± 7.8 pp, low confidence |
| LFM2.5-VL-1.6B-Extract | 0.0% | 74.7% | estimated ± 7.8 pp, low confidence |
| Ling 2.6 Flash | 0.0% | 74.7% | estimated ± 7.8 pp, low confidence |
| Ling 3.0 Tiny | 0.0% | 74.7% | estimated ± 7.8 pp, low confidence |
| Llama 3.1 405B | 0.0% | 74.7% | estimated ± 7.8 pp, low confidence |
| Llama 3.1 70B | 0.0% | 74.7% | estimated ± 7.8 pp, low confidence |
| Llama 3.3 70B | 0.0% | 74.7% | estimated ± 7.8 pp, low confidence |
| Llama 4 Maverick | 0.0% | 74.7% | estimated ± 7.8 pp, low confidence |
| Llama 4 Scout | 0.0% | 74.7% | estimated ± 7.8 pp, low confidence |
| Magistral Small | 0.0% | 74.7% | estimated ± 7.8 pp, low confidence |
| Mercury 2.5 | 0.0% | 74.7% | estimated ± 7.8 pp, low confidence |
| Mistral Large 2 | 0.0% | 74.7% | estimated ± 7.8 pp, low confidence |
| Mistral Medium 3 | 0.0% | 74.7% | estimated ± 7.8 pp, low confidence |
| Mistral Medium 3.5 128B | 0.0% | 74.7% | estimated ± 7.8 pp, low confidence |
| Mistral Small 3.1 | 0.0% | 74.7% | estimated ± 7.8 pp, low confidence |
| Nova Pro | 0.0% | 74.7% | estimated ± 7.8 pp, low confidence |
| Phi-4 | 0.0% | 74.7% | estimated ± 7.8 pp, low confidence |
| Phi-4-reasoning | 0.0% | 74.7% | estimated ± 7.8 pp, low confidence |
| Qwen3.5-4B | 0.0% | 74.7% | estimated ± 7.8 pp, low confidence |
| Qwen3-Omni-30B-A3B-Instruct | 0.0% | 74.7% | estimated ± 7.8 pp, low confidence |
| Qwen3-Omni-30B-A3B-Thinking | 0.0% | 74.7% | estimated ± 7.8 pp, low confidence |
| Sarvam 105B | 0.0% | 74.7% | estimated ± 7.8 pp, low confidence |
| Solar Pro 2 | 0.0% | 74.7% | estimated ± 7.8 pp, low confidence |
| Solar Pro 3 | 0.0% | 74.7% | estimated ± 7.8 pp, low confidence |
| Ultravox v0.6 Llama 3.3 70B | 0.0% | 74.7% | estimated ± 7.8 pp, low confidence |
Fitted on
| Model | CritPt | ARC-AGI-1 |
|---|---|---|
| Claude Fable 5 | 28.6% | 98.5% |
| Claude Fable 5.1 | 29.7% | 97.5% |
| Claude Haiku 5.5 | 18.9% | 89.0% |
| Claude Opus 4.5 Thinking | 4.6% | 80.0% |
| Claude Opus 4.6 (Adaptive) | 12.6% | 93.0% |
| Claude Opus 4.8 | 20.9% | 92.5% |
| Claude Opus 5 | 29.1% | 97.5% |
| Claude Opus 5.5 | 31.7% | 97.5% |
| Claude Sonnet 4.6 | 0.9% | 86.0% |
| DeepSeek V4 Flash 0731 | 16.6% | 89.0% |
| DeepSeek V4 Pro | 18.0% | 90.0% |
| Gemini 3.5 Flash | 13.1% | 92.5% |
| Gemini 3.6 Flash | 10.6% | 91.2% |
| Gemini 3.7 Flash | 14.3% | 95.5% |
| Gemini 3.8 Flash | 18.3% | 98.5% |
| Gemini 3 Pro | 9.1% | 75.0% |
| Gemini 4 Argon | 27.1% | 98.5% |
| GPT-5.1 | 4.9% | 72.8% |
| GPT-5.2 | 11.6% | 86.2% |
| GPT-5.4 | 23.4% | 93.7% |
| GPT-5.4 mini | 10.0% | 63.7% |
| GPT-5.4 nano | 9.3% | 51.5% |
| GPT-5.4 Pro | 30.0% | 94.5% |
| GPT-5.5 | 27.1% | 95.0% |
| GPT-5.5 Pro | 30.6% | 95.0% |
| GPT-5.6 Luna | 20.6% | 88.0% |
| GPT-5.6 Sol | 32.3% | 96.5% |
| GPT-5.6 Terra | 30.0% | 96.5% |
| GPT-6.1 Sol | 31.7% | 96.5% |
| GPT-6 Astra | 31.7% | 97.5% |
| GPT-6 Luna | 19.4% | 86.7% |
| GPT-6 Sol | 30.9% | 95.5% |
| Grok 4.5 | 15.4% | 85.7% |
| Grok 4.6 | 17.1% | 87.0% |
| Grok 4.7 | 17.7% | 89.5% |
| Inkling-Small | 8.3% | 84.0% |
| Kimi K3 | 23.4% | 94.5% |