MMOOC - Misleading Premise (Open-Ended VQA): leaderboard
Metric: Shifted in-context score (%; mean of accuracy on the answerable Misleading Premise Open-Ended VQA questions and answer rationality, the mean 0-1 judge score of the answer's correctness and evidence consistency in steps of 0.25; judge scores averaged over GPT-5.6, Claude Opus 5 and DeepSeek-V4-Pro). Source: arxiv.org. Saturation forecast: Around December 2026. 18 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Qwen 3.5 27B | 90.5 |
| 2 | Qwen 3.5 122B A10B | 90.25 |
| 3 | Llama 4 Maverick | 77.5 |
| 4 | GPT-4o | 76.75 |
| 5 | Ministral 3 8B | 76.75 |
| 6 | Gemma 4 31B | 75.75 |
| 7 | O1 | 74.75 |
| 8 | Gemma 4 26B | 74.25 |
| 9 | O3 | 71 |
| 10 | Ministral 3 14B | 65.25 |
| 11 | InternVL3-8B | 62.5 |
| 12 | Gemini 3.1 Pro (Preview) | 61.5 |
| 13 | Claude Opus 4.6 | 21.75 |
Interactive version: theaggregate.ai/benchmark?slug=mmooc-misleading-premise-open-ended-vqa · How It Works · Data refreshed daily, snapshot 2026-09-29.