MMOOC - Missing Knowledge and Background (Open-Ended VQA): leaderboard
Metric: Out-of-context score (%; mean of the refusal rate, the share of the Missing Knowledge and Background Open-Ended VQA questions, which the image and context cannot answer, that the model correctly declines, and refusal rationality, the mean 0-1 judge score of its abstention reasoning 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 2030. 18 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemma 4 26B | 80 |
| 2 | Gemma 4 31B | 70.5 |
| 3 | Llama 4 Maverick | 58 |
| 4 | Ministral 3 8B | 58 |
| 5 | GPT-4o | 57.75 |
| 6 | O1 | 53.75 |
| 7 | Claude Opus 4.6 | 46.5 |
| 8 | Qwen 3.5 122B A10B | 41.5 |
| 9 | O3 | 41.25 |
| 10 | Qwen 3.5 27B | 39 |
| 11 | Ministral 3 14B | 36.5 |
| 12 | InternVL3-8B | 32 |
| 13 | Gemini 3.1 Pro (Preview) | 20.25 |
Interactive version: theaggregate.ai/benchmark?slug=mmooc-missing-knowledge-and-background-open-ended-vqa · How It Works · Data refreshed daily, snapshot 2026-09-29.