ContrAR (Single Frame) - Smart Apartment: leaderboard
Metric: Detection accuracy (%) for contradictory virtual content attacks in AR videos (90 smart apartment videos; 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: Estimated already saturated. 11 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5 | 94.44 |
| 2 | Gemini 2.5 Flash | 88.89 |
| 3 | Gemini 2.5 Pro | 86.67 |
| 4 | GPT-4.1 | 84.44 |
| 5 | GPT-4o | 77.78 |
| 6 | Claude Sonnet 4.5 | 64.44 |
| 7 | Qwen 2.5 VL 72B Instruct | 62.22 |
| 8 | Grok 4 | 61.11 |
| 9 | Qwen 2.5 VL 7B Instruct | 57.78 |
| 10 | Claude Haiku 4.5 | 48.89 |
Interactive version: theaggregate.ai/benchmark?slug=contrar-single-frame-smart-apartment · How It Works · Data refreshed daily, snapshot 2026-10-07.