VIA-Bench (LLM Judge) - Motion Illusions: leaderboard
Metric: Accuracy (%) on the motion illusions questions, answer letter read from the response by a GPT-4.1-mini judge; VIA-Bench's 1,004 human-verified multiple-choice questions on illusory and anomalous images, every option set including a never-correct 'Not Sure' option; zero-shot with an answer-format instruction, temperature 0.8 where supported (model defaults otherwise), mean of five runs; higher is better. Source: arxiv.org. Saturation forecast: Estimated already saturated. 24 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3 Pro (Preview) | 99.36 | #64 |
| 2 | Qwen 3 VL 30B A3B (Thinking) | 96.28 | #338 (Qwen 3 VL 30B A3B) |
| 3 | Qwen 3 VL 235B A22B (Thinking) | 95.51 | #228 (Qwen 3 VL 235B A22B) |
| 4 | O4 Mini | 94.87 | #172 |
| 5 | O3 | 90.51 | #121 |
| 6 | Claude Opus 4.1 (20250805) | 90.26 | #126 |
| 7 | Qwen 2.5 VL 7B Instruct | 78.97 | #643 |
| 8 | Qwen 2.5 VL 72B Instruct | 56.03 | #364 |
| 9 | GPT-5 Chat | 47.05 | #233 |
| 10 | Gemini 2.5 Pro | 46.54 | #145 |
| 11 | Claude Sonnet 4 (20250514) | 41.28 | #211 |
| 12 | GPT-4o ChatGPT | 35.51 | |
| 13 | GPT-5 Mini | 25.13 | #176 |
| 14 | Claude 3.5 Sonnet (20241022) | 20.38 | #286 |
| 15 | GPT-4o (2024-11-20) | 20.13 | #369 |
No result here: #3 Claude Opus 5.5, #5 GPT-6 Astra, #8 Claude Fable 5.1.
Interactive version: theaggregate.ai/benchmark?slug=via-bench-llm-judge-motion-illusions · How It Works · Data refreshed daily, snapshot 2026-10-11.