SDGBiasBench - MCQ (No Context): leaderboard
Metric: Accuracy (%; three-option ordinal questions on SDG indicators from satellite and street-level imagery, no structured context, image and question only). Source: arxiv.org. Saturation forecast: Estimated already saturated. 13 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Qwen 2.5 VL 7B | 53 |
| 2 | GPT-4o | 47 |
| 3 | InternVL3-8B | 47 |
| 4 | Gemini 2.0 Flash | 44 |
| 5 | Claude 3.5 Sonnet | 42 |
| 6 | InternVL3-14B | 40 |
Interactive version: theaggregate.ai/benchmark?slug=sdgbiasbench-mcq-no-context · How It Works · Data refreshed daily, snapshot 2026-09-26.