Developer Ryan Shea launched AI IQ (aiiq.org), a platform that places cutting-edge AI models on the human IQ bell curve, providing a single number to answer "How smart is this model?" Currently, data on the site shows for key tested models:
- GPT-5.5: 136 (top)
- Claude Opus 4.7 and Gemini 3.1 Pro: tied at 132
- Grok 4.3: 125
- Kimi K2.6: 122
- DeepSeek V4 Pro and Muse Spark: tied at 117
- Qwen3.6: 108
The platform's algorithm scrapes raw scores from 12 public benchmarks, converts them to an implied IQ using calibrated difficulty curves, and averages across four dimensions: abstract reasoning, math reasoning, coding reasoning, and academic reasoning. Missing dimensions are conservatively filled so models don't benefit from incomplete data. Since all underlying data comes from existing benchmarks, AIIQ doesn't run new tests — its core value is translating scattered scores into an intuitive scale.
Beyond the overall IQ ranking, the site offers cross-views. Users can filter by IQ vs. cost in a chart, or see the evolution slope of different labs on the "Frontier IQ Timeline." The platform also overlays EQ scores from EQ-Bench, measuring how well a model "minds its manners."