Automated page. Written and translated automatically by AI and checked against the sources by software; no human editor. SI-IQ is an index built from public benchmark results, not an IQ test. How it works
How smart is each AI model?
SI-IQ combines 13 independent benchmarks from 4 evaluators into one IQ-style number per model. Every input score links to its source, and the leaderboard refreshes every hour.
SI-IQ is a statistical index on an IQ-style scale: 100 is the average of a reference group of AI models released in 2025, and 15 points is one standard deviation of that group. It is not an IQ test and does not measure human-like intelligence.
Top 10 right now
| # | Model | SI-IQ |
|---|---|---|
| 1 | GPT-6 AstraOpenAI | 150146–153 |
| 2 | GPT-6.1 SolOpenAI | 148144–152 |
| 3 | Claude Opus 5.5Anthropic | 144141–148 |
| 4 | Claude Sonnet 5.5Anthropic | 140137–143 |
| 5 | GPT-6 SolOpenAI | 140137–144 |
| 6 | Gemini 4 ArgonGoogle DeepMind | 140137–143 |
| 7 | GPT-5.6 SolOpenAI | 140137–143 |
| 8 | Claude Fable 5.1Anthropic | 140136–143 |
| 9 | GPT-5.5OpenAI | 138135–141 |
| 10 | Claude Fable 5Anthropic | 138134–141 |