TrackingAI: leaderboard
Monitors AI chatbots over time, tracking political ideology (Political Compass) and IQ scores (Mensa tests). Reveals trends and biases across regular re-testing.
Metric: IQ Test Score (%). Source: trackingai.org. Source: trackingai.org. 40 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Fable 5 | 98.04 | #16 |
| 2 | Claude Opus 5 (Max) | 98.04 | #12 (Claude Opus 5) |
| 3 | GPT-5 Pro | 96.08 | #81 |
| 4 | Kimi K3 | 93.75 | #30 |
| 5 | Grok 4 | 92.16 | #181 |
| 6 | Gemini 3 Pro | 92.16 | #80 |
| 7 | Claude Opus 5 | 92.16 | #12 |
| 8 | GPT-5 (Thinking) | 90.2 | #98 (GPT-5) |
| 9 | O3 Pro | 90.2 | #116 |
| 10 | Claude Opus 4 (Thinking) | 86.27 | #163 (Claude Opus 4) |
| 11 | O3 | 84.31 | #124 |
| 12 | Grok 4.5 | 84.31 | #48 |
| 13 | Claude Sonnet 5 | 82.35 | #57 |
| 14 | O1 Pro | 80.39 | #228 |
| 15 | GPT-5 | 78.43 | #98 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=trackingai · How It Works · Data refreshed daily, snapshot 2026-10-20.