ContrAR (Multiple Frames) - 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), first, middle and last frames given (multi-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-4.1 | 93.33 |
| 2 | GPT-5 | 92.22 |
| 3 | GPT-4o | 86.67 |
| 4 | Gemini 2.5 Pro | 82.22 |
| 5 | Gemini 2.5 Flash | 76.67 |
| 6 | Claude Sonnet 4.5 | 67.78 |
| 7 | Qwen 2.5 VL 72B Instruct | 58.89 |
| 8 | Grok 4 | 57.78 |
| 9 | Claude Haiku 4.5 | 55.56 |
| 10 | Qwen 2.5 VL 7B Instruct | 54.44 |
Interactive version: theaggregate.ai/benchmark?slug=contrar-multiple-frames-smart-apartment · How It Works · Data refreshed daily, snapshot 2026-10-07.