Benchmark
TrackingAI IQ (Offline Test)
Higher scores are better.
- Models measured
- 39
- Scores recorded
- 67
- Sources
- 1
- IQ weight
- 1.38
Score distribution
- 97.1
- Mean
- 103
- Median
- 26
- Std dev
- 57
- 10th percentile
- 123
- 90th percentile
- 44
- Min
This benchmark has a weight of 1.38 for chat models, measured over 39 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.
Leaderboard
| # | Model | Score |
|---|---|---|
| 51 | Llama 4 Maverick· Tracking AI data | 77 |
| 52 | Llama 4 Maverickvision· Tracking AI data | 77 |
| 53 | Manus· Tracking AI data | 77 |
| 54 | Grok 3vision, think· Tracking AI data | 70 |
| 55 | Claude Opus 4.8vision· Tracking AI data | 64 |
| 56 | Grok 2· Tracking AI data | 64 |
| 57 | Claude 3.5 Sonnet· Tracking AI data | 57 |
| 58 | Claude Sonnet 4.6vision· Tracking AI data | 57 |
| 59 | GPT-4 Turbo· Tracking AI data | 57 |
| 60 | o4 Mini· Tracking AI data | 57 |
| 61 | o4 Minivision· Tracking AI data | 57 |
| 62 | OpenAI o3vision· Tracking AI data | 57 |
| 63 | Claude 3 Opusvision· Tracking AI data | 50 |
| 64 | GPT 4overbal· Tracking AI data | 50 |
| 65 | Llama 3.3 70B Instruct· Tracking AI data | 50 |
| 66 | O1 Provision· Tracking AI data | 50 |
| 67 | GPT 4ovision· Tracking AI data | 44 |
Showing 51–67 of 67 results.