ReaLM - PA-Attack: leaderboard
Metric: Attack success rate (%; share of adversarial cases where the model gives an incorrect, task-relevant answer) under PA-Attack (untargeted out-of-distribution prototype attack, L-inf 8/255) on the 832 ReaLM physical-world cases (driving, manipulation, grasping, physics, egocentric, diagnostics and scene QA; image, question and ground-truth answer), black-box one-shot evaluation at temperature 0, a Qwen3-8B judge extracting the answer; lower is better. Source: arxiv.org. Saturation forecast: Around 2028. 13 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Qwen 3.5 27B | 38.1 |
| 2 | Qwen 3.5 122B A10B | 39.4 |
| 3 | GPT-4.1 Mini | 40.6 |
| 4 | Qwen 3.6 Flash | 41.7 |
| 5 | Gemini 3 Flash | 41.9 |
| 6 | Qwen 3 VL 32B | 42.3 |
| 7 | Qwen 3.5 9B | 42.8 |
| 8 | Kimi K2.5 | 43 |
| 9 | Seed 2.0 Mini | 45.4 |
Interactive version: theaggregate.ai/benchmark?slug=realm-pa-attack · How It Works · Data refreshed daily, snapshot 2026-09-29.