ContrAR (Single Frame): leaderboard
Metric: Detection accuracy (%) for contradictory virtual content attacks in AR videos (all 312 videos of five AR applications; half the videos carry a contradiction, a balanced yes/no decision), one middle frame of each video given (single-frame prompting); mean of three runs; higher is better. Source: arxiv.org. Saturation forecast: Around January 2027. 11 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5 | 88.14 |
| 2 | Gemini 2.5 Pro | 83.97 |
| 3 | GPT-4.1 | 82.05 |
| 4 | Gemini 2.5 Flash | 79.81 |
| 5 | GPT-4o | 79.17 |
| 6 | Grok 4 | 68.27 |
| 7 | Claude Sonnet 4.5 | 63.78 |
| 8 | Qwen 2.5 VL 72B Instruct | 60.26 |
| 9 | Qwen 2.5 VL 7B Instruct | 58.01 |
| 10 | Claude Haiku 4.5 | 53.85 |
Interactive version: theaggregate.ai/benchmark?slug=contrar-single-frame · How It Works · Data refreshed daily, snapshot 2026-10-07.